Plants and Bioenergy

Advances in Plant Biology 4
Maureen C.McCann
Marcos S.Buckeridge
Nicholas C.Carpita Editors
Plants and
BioEnergy
Advances in Plant Biology
Volume 4
Series Editor
John J. Harada, Davis, CA, USA
For further volumes:

http://www.springer.com/series/8047
Maureen C. McCann Marcos S. Buckeridge
Nicholas C. Carpita
Editors
Plants and BioEnergy
13
Editors
Maureen C. McCann Nicholas C. Carpita
Department of Biological Sciences Department of Botany and Plant Pathology
Purdue University Purdue University
West Lafayette West Lafayette
IN, USA IN, USA
Marcos S. Buckeridge
Department of Botany
Institute of Biosciences
University of Campinas
Sao Paolo
Brazil
ISBN 978-1-4614-9328-0 ISBN 978-1-4614-9329-7 (eBook)

DOI 10.1007/978-1-4614-9329-7
Springer New York Heidelberg Dordrecht London
Library of Congress Control Number: 2013953273
Springer Science+Business Media New York 2014

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or
information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed. Exempted from this legal reservation are brief excerpts
in connection with reviews or scholarly analysis or material supplied specifically for the purpose of
being entered and executed on a computer system, for exclusive use by the purchaser of the work.
Duplication of this publication or parts thereof is permitted only under the provisions of the Copyright
Law of the Publishers location, in its current version, and permission for use must always be obtained
from Springer. Permissions for use may be obtained through RightsLink at the Copyright Clearance
Center. Violations are liable to prosecution under the respective Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of
publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for
any errors or omissions that may be made. The publisher makes no warranty, express or implied, with
respect to the material contained herein.
Printed on acid-free paper
Springer is part of Springer Science+Business Media (www.springer.com)

Preface
A nations vision for developing renewable and sustainable energy resources is

typically propelled by three driverssecurity, cost, and environmental impact.
The U.S. currently accounts for one-quarter of the worlds total oil consumption.
Technology improvements in the recovery of oil from shale, production of hybrid
and electric vehicles, and light vehicle fuel efficiencies have reduced but far from
eliminated dependence on foreign oil imports. At the same time, Brazil, because
of its embrace of ethanol as an alternative liquid fuel in the 1970s, is today energy
independent. Issues of energy security are compounded by increased demand from
emerging economies and the supply of that demand from politically unstable parts
of the world. Economic growth and development worldwide depend increasingly
on secure supplies of reliable, affordable, and clean energy. As easily accessible
reserves of oil become exhausted, the energy return on energy invested, currently
a ratio of 30:1, will decrease, driving up costs for the consumer, not just of liquid
transportation fuels, but of all of the oil-based chemicals and materials supplied
by the petrochemical industry. Agriculture itself is an oil-intensive enterprise, with
about 2 % of total oil consumption used directly in farm vehicles or indirectly for
mechanized processing. As carbon dioxide levels reach an unprecedented 400
ppm, there is an unequivocal imperative to mitigate greenhouse gas emissions
by decreasing fossil fuel consumption and transition to carbon-neutral or carbon-
negative fuels as well as improving efficiency of fuel use.
It was with the urgency conferred by these three drivers that the American
Society of Plant Biologists convened the First Pan-American Congress on Plants
and BioEnergy in June, 2008, in Mrida, Mexico. This congress was designed
to initiate Pan-American research collaborations in energy biosciences and to
showcase advances in the development of new energy crop plants, their genetic
improvement based on new knowledge of plant growth and development, their
fit into regional environments, and the development of a sustainable energy
agriculture. Subsequent biennial meetings, one in Brazil and another in the US,
have served to connect advances in second and third generation of biofuels with the
realities of economic success and sustainability. This edition encompasses specific
examples of progress to this goal yet keeps in perspective the realities of the eco-
nomic drivers and pressures that govern the translation of scientific success into a
commercial success.
v
vi Preface
In Part I, Social and Economic Impacts of a Bioenergy Agriculture, we

begin with Patricia Guardabassi and Jos Goldembergs overview of the pros-
pects of global ethanol production in developing countries and the relevant social
and economic issues unique to their environments. Jeremy Woods and Nicole
Kalas extend this theme, drawing on the lessons of biofuels policy development
concerning direct and indirect land use over the last decade to inform energy
policies that will drive sustainable land use. Wally Tyner and Farzad Taheripour
discuss the uncertainties, policy options, and land-use impacts in moving from
ethanol to advanced fuels. Finally, Gal Hochman and David Zilberman discuss the
implementation and economics of algal farming and how economic success hinges
on high-value bio-products generation.
In Part II, Biomass Feedstocks, we explore the breadth of bioenergy crops
and the progress that is being made to introduce them into the agricultural
landscape, the underlying biology of bioenergy plants, and new ideas to enhance
biomass yield and quality for the energy crops of the future. Andrew Jakubowski
and Michael Casler show that improved and locally collected ecotypes of
switchgrass, big bluestem, and Indiangrass can coexist on the landscape and
help to jumpstart sustainably the shift to a bioenergy-based economy. Cynthia
Damasceno, Robert Schaffert, and Ismail Dweikats article considers how to mine
the vast genetic diversity in the sorghum genome and its advantages as an annual
crop for use in both tropical and temperate biomes. Angela Karp and her col-
leagues explore the challenges and prospects for integrative approaches to improve
woody biomass species, such as willow, as lignocellulosic feedstocks. Two per-
spectives consider oil production platforms for advanced biofuels, biodiesel, and
bio-based products. Umidjon Iskandarov, Hae Jin Kim, and Edgar Cahoon present
the advantages of Camelina as an emerging drought-resistant oilseed suitable for
marginal lands that are tractable to genetic improvement, and Janaina Meyer and
Antonio Salatino, present their ideas for Brazils contribution to biodiesel with
palm. To conclude this section, Ahmed Faik, Nan Jiang, and Mick Held present a
thorough update on our present knowledge of the biochemistry of xylan synthesis,
with particular emphasis on the unique aspects of synthesis in grasses. Catherine
Rayon, Anna Olek, and Nick Carpita close this section with a perspective on the
complexities of cellulose biosynthesis that suggests strategies for how cellulose
might be designed for improved bioenergy feedstocks.
In Part III, Biomass Conversion Technologies we explore the culmination of the
technologies that drive the ethanol industry and the promise for the efficient conver-
sion of biomass into energy-dense liquid fuels and high value co-products. Harry
Gilbert begins with an extensive review of how novel enzyme repertories are devel-
oped for the efficient deconstruction of plant biomass tailored for the bioenergy
industry. Adriana Grandis and her colleagues in the Laboratory of Marcos Buckeridge
extend this strategy in their perspective on exploiting natural plant cell wall degrada-
tion systems to improve bioethanol production. Rebecca Garlock Ong and colleagues
in the laboratory of Bruce Dale present a comprehensive review on how knowl-
edge of the fine structure of the plant cell wall informs better design principles for
the biorefinery. Ken Reardon gives some guiding principles for the lignocellulosic
Preface vii
refineries of the future for how both the economics and the environmental impacts
of biofuel production could be improved by developing processes to obtain a wider
range of chemicals with higher value from biomass. The edition concludes with
two articles that explore the early successes in the direct conversion of lignocel-
lulosic biomass into advanced biofuels. Basudeb Saha, Nathan Mosier, and Mahdi
Abu-Omar show the path to catalytic dehydration of lignocellulosic-derived xylose to
furfural, while Joe Bozell identifies pathways for the catalytic oxidation of lignin for
the production of low molecular weight aromatics.
In this volume, we bring together perspectives from a wide range of disciplines,
recognizing that the grand challenge of displacing a century of global dependence
on oil requires a new research paradigm, a Manhattan Project for the twenty-first
century. Production of carbon-neutral reliable, affordable biofuels for a growing
population in a manner that does not compromise food and feed production takes a
community that is fully engaged, committed, and international in scope. The suc-
cess of that community will be measured in our contributions to climate security,
economic growth, and self-sufficient energy production for our nations.
Maureen C. McCann
Marcos S. Buckeridge
Nicholas C. Carpita
Contents
Part I Economics of Bioenergy
1 The Prospects of First Generation Ethanol

in Developing Countries. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Patricia Guardabassi and Jos Goldemberg
2 Can Energy Policy Drive Sustainable Land Use? Lessons

from Biofuels Policy Development Over the Last Decade. . . . . . . . . . . 13
Jeremy Woods and Nicole Kalas
3 Advanced Biofuels: Economic Uncertainties, Policy Options,

and Land Use Impacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Wallace E. Tyner and Farzad Taheripour
4 Algae Farming and Its Bio-Products. . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

Gal Hochman and David Zilberman
Part II Biomass Biology
5 Regional Gene Pools for Restoration, Conservation,

and Genetic Improvement of Prairie Grasses . . . . . . . . . . . . . . . . . . . . 67
Andrew R. Jakubowski and Michael D. Casler
6 Mining Genetic Diversity of Sorghum as a Bioenergy Feedstock. . . . . 81

Cynthia M. B. Damasceno, Robert E. Schaffert and Ismail Dweikat
7 Genetics, Genomics and Crop Modelling: Integrative Approaches

to the Improvement of Biomass Willows. . . . . . . . . . . . . . . . . . . . . . . . . 107
Angela Karp, Goetz M. Richter, Ian F. Shield and Steven J. Hanley
ix
x Contents
8 Camelina: An Emerging Oilseed Platform for Advanced Biofuels

and Bio-Based Materials. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
Umidjon Iskandarov, Hae Jin Kim and Edgar B. Cahoon
9 Perspectives in Brazil of the Contribution of Palm Trees

to Biodiesel Production. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
Janaina M. Meyer and Antonio Salatino
10 Xylan Biosynthesis in Plants, Simply Complex . . . . . . . . . . . . . . . . . . . 153

Ahmed Faik, Nan Jiang and Michael A. Held
11 Towards Redesigning Cellulose Biosynthesis

for Improved Bioenergy Feedstocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
Catherine Rayon, Anna T. Olek and Nicholas C. Carpita
Part III Biomass Processing
12 Developing Novel Enzyme Repertoires for the Efficient

Deconstruction of Plant Biomass Tailored for the Bioenergy
Industry. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
Harry J. Gilbert
13 Using Natural Plant Cell Wall Degradation Mechanisms

to Improve Second Generation Bioethanol. . . . . . . . . . . . . . . . . . . . . . . 211
Adriana Grandis, Amanda P. de Souza, Eveline Q. P. Tavares
and Marcos S. Buckeridge
14 Linking Plant Biology and Pretreatment: Understanding

the Structure and Organization of the Plant Cell Wall
and Interactions with Cellulosic Biofuel Production. . . . . . . . . . . . . . . 231
Rebecca Garlock Ong, Shishir P. S. Chundawat, David B. Hodge,
Sai Keskar and Bruce E. Dale
15 Lignocellulosic Biorefineries: Concepts and Possibilities . . . . . . . . . . . 255

Kenneth F. Reardon
16 Catalytic Dehydration of Lignocellulosic Derived Xylose

to Furfural. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
Basudeb Saha, Nathan S. Mosier and Mahdi M. Abu-Omar
17 Catalytic Oxidation of Lignin for the Production of Low Molecular

Weight Aromatics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
Joseph J. Bozell
Part I
Economics of Bioenergy
Chapter 1
The Prospects of First Generation Ethanol
in Developing Countries
Patricia Guardabassi and Jos Goldemberg
AbstractThere are great perspectives to the development of second-generation

technologies to biofuels production, nevertheless its production in large scale is
depending on a technological breakthrough to become feasible. The production of
ethanol from sugarcane based on first generation technology has evolved in the
last decades; however gains of productivity can still be achieved. Latin American
and African countries have suitable conditions to the growth of sugarcane. Many
of these countries are highly dependent on fossil fuels imports. Thus, the introduc-
tion of ethanol blends can reduce the consumption of fossil fuels, while creating
jobs and developing local industry. Notwithstanding, first generation ethanol can
still contribute to developed countries, especially US and European countries, to
commit with biofuels use mandates. The aim of this chapter is to present the state
of the art of ethanol production in Latin America and African countries, identify-
ing the main obstacles to the development and discussing policies that could be
implemented to overcome such barriers.
Keywords Biofuels First generation Developing countries Barriers
P. Guardabassi
Sustainability Science Program, Kennedy School of Government, Harvard University,
79 John F. Kennedy Street, Cambridge, MA 02138, USA
e-mail: Patricia_Guardabassi@hks.hrvard.edu
J. Goldemberg (*)
Institute of Eletrotechnics and Energy, University of So Paulo, Av. Prof. Luciano Gualberto,
1289, 05508-010 So Paulo/SP, Brazil
e-mail: goldemb@iee.usp.br
M. C. McCann et al. (eds.), Plants and BioEnergy, Advances in Plant Biology 4, 3

DOI: 10.1007/978-1-4614-9329-7_1, Springer Science+Business Media New York 2014
4 P. Guardabassi and J. Goldemberg
1.1Introduction
The economic growth of developing countries and the maintenance of consump-

tion patterns in developed countries continuously increase worlds energy con-
sumption. Concomitantly, the depletion of oil reserves has been observed as well
as the impacts of climate change caused by human actions, especially due fossil
fuels consumption. In the specific case of energy, this topic is of special impor-
tance since the projections indicate, in a scenario based on current policies for the
energy sector, a growth of the worlds energy consumption by 47% between the
years 2008 and 2035 (OECD/IEA 2010), based on increased use of coal and nat-
ural gas. In this context it is necessary to develop alternative sources of energy
and modern technologies that can replace fossil fuels and mitigate greenhouse gas
(GHG) emissions.
The most promising alternatives are those based on renewable sources of
energy such as solar panels and wind turbines. However for the transportation
sector, which accounted for 13% of greenhouse gases global emissions of 2004
(IPCC 2007), biofuels are the only worldwide commercially available option,
since electric vehicles or hydrogen still demand technological development. Thus,
biofuels are gaining an increasingly importance to reduce GHG emissions as well
as the dependence of fossil fuels. In view of the impacts from the extensive use of
fossil fuels, and the opportunity to reduce their emissions through the adoption of
renewable energy in their energy matrixes, developed countries have established
targets for use of clean fuels (Goldemberg 2007).
The most ambitious one was established by the European Union, whose goals
include a 20% share of renewable energies in overall Community consumption
and, in the transportation sector, at least 10% biofuels by 2020, equivalent to
14billion liters. However, alleging aiming to reduce the impact of first genera-
tion biofuels on food prices, the EU cut by a half its biofuels target. Within this
bioenergy program the European Parliament and the European Union Council
established the European Union Directive 28/2009 on the promotion of the use of
energy from renewable sources. This Directive sets sustainability criteria that must
be accomplished by countries willing to provide biofuels to the European Union
country members. In the case of European Union, biofuels are expected to provide
a great contribution in the achievement of renewable energy use target and GHG
emissions in transport sector (European Expert Group on Future Transport Fuels
2011).
Another initiative that will also be central for the global biofuels market due
to its dimensions was taken by the U.S. government in the Energy Independence
and Security Act (EISA). Established in 2007, it set a minimum consumption in
the country of 45billion liters of biofuels by 2010, reaching 136billion liters in
2022. The legislation identifies three types of biofuels, which should account for
about 60% of this volume (equivalent to 88billion liters), they are: cellulosic
ethanol, biomass diesel and other advanced. To be classified as advanced
biofuels must reduce by 50% GHG emissions on a life cycle basis, compared to
1 The Prospects of First Generation Ethanol in Developing Countries 5
gasoline. The levels of GHG emissions reduction for biofuels were adopted by the
Environmental Protection Agency (EPA), and according to the Agencys calcula-
tions, Brazilian ethanol reduces GHG emissions by 61%. In summary, European
and American initiatives together are responsible for the demand of 150billion
liters of ethanol in the year 2022.
However, due to edaphoclimatic conditions and restrictions on land availability,
many countries do not produce volumes of biofuels sufficient to supply the domes-
tic market. Consequently, the international trade of biofuels has been growing and
more producing countries are likely to be part of it. Ethanol can be produced from
a series of feedstocks; however its access to American and European markets will
depend on its GHG emissions balance (Worldwatch Institute 2007). Among dif-
ferent raw materials to produce ethanol, sugarcane is the most effective in GHG
emissions reduction, because small amounts of fossil fuels are used in its pro-
duction chain. According to (Smeets et al. 2007), for the year 2050, Sub-Saharan
Africa and Latin America and the Caribbean regions have the greatest potential for
production of agricultural residues that could be used as bioenergy feedstock.
The Global Agro-Economic Zones (GAEZ) system developed by FAO, in
conjunction with the International Institute for Applied Systems Analysis (IIASA),
identified and quantified areas with potential to produce raw materials for biofuels
based on climate and soil conditions. In the case of rain-fed sugar cane, GAEZ
identified 135million suitable or very suitable hectares and 130million hectares
moderately suitable hectares (of which only 22million are currently planted with
sugar cane). For unprotected areas the overall potential is estimated at 87million
hectares suitable or very suitable, of which 26million hectares are located Africa
and 54million hectares South America (Fischer et al. 2009). The regions with
greater aptitude for the cultivation of rain-fed sugar cane are located in South and
Central America, the Caribbean, Central Africa and some countries in West and
South Africa, South Indian and Southeast Asia. Although suitable for the produc-
tion of biofuels these regions produce small amounts of ethanol.
From the worlds total production of ethanol, in 2010, of 86.8billion liters;
United States and Brazil were responsible for 76.2billion; and despite its poten-
tial the African continent produced only 0.16billion liters and Latin America
and Caribbean 2.6billion liters (RFA 2011). In fact, there are barriers that must
be overcome in order to allow the development of a biofuels production sector.
In order to surpass such barriers, it is necessary to settle a legal framework that
ensures an appropriate institutional condition, economically attractive to investors
and to promote the production minimizing environmental and social impacts.
1.2The State of the Art of Ethanol Production
Ethanol production has been based in the so-called first generation technolo-
gies either by direct fermentation (in the case of sugar cane) or saccharification of
starch (in the case of corn and wheat) followed by fermentation. New technologies,
or second generation technologies, to produce biofuels include the enzymatic con-

version of lignocellulosic material. Despite huge amounts invested and the demand
created by US Renewable Fuel Standard, there are no commercial plants operat-
ing yet. According to the Advanced Ethanol Council Advanced Ethanol Council
(AEC) (2012), industry is reaching commercial deployment phase, however the
high capital risk from OPEC-induced price distortions, constrained blending mar-
kets and policy uncertainty continues to slow down the rate of deployment.
Currently, the US is the largest ethanol producer with an annual amount of
49billion liters, using corn as feedstock, followed by Brazil, which uses sugarcane
as feedstock, producing 28billion liters (REN21 2012). So far, sugarcane etha-
nol has proved to be the most competitive raw material in use, due to its positive
energy balance, and consequently positive reduction of greenhouse gas emissions
on a life cycle basis, the lower costs of production and higher production yields
when compared to corn (Goldemberg, The Brazilian biofuels industry 2008).
Therefore, it is plausible to extend a successful experience, such as the Brazilian
ethanol program, to other developing countries.
1.3Existing Mandates to Ethanol Production and Use
1.3.1Africa
Ethanol is produced in Africa to replace gasoline since de 1970s, in countries

highly dependent on imports, such as Zimbabwe, Malawi and Kenya. Malawi pro-
duces ethanol since 1982 and uses E10 blends. The country is running tests with
ethanol dedicated engines. The local production reaches 18million liters, half of
it domestically consumed and the rest exported to African countries. There are two
distilleries operating in the country with a total producing capacity of 32million
litres. Government aims to estimulate the production of sugarcane in order to use
this idle capacity (Janssen and Rutz 2012).
Other nations are introducing policies aiming to leverage the production of
biofuels aiming to increase energy security due to energy matrix diversification
and reduction of fossil fuels imports. Hence, such policies are also instruments
to promote the development of rural areas, through job creation, income genera-
tion, local industry expansion and investments in infrastructure. South Africa is
the largest market in the region. The country established The Biofuels Industrial
Strategy of the Republic of South Africa, in December 2007, which aims to
develop a biofuels industry that could supply the domestic market of 2% etha-
nol blends (equivalent to 400million liters per year) within 5years. Sugarcane
and sugar beet are edible crops to ethanol production, and sunflower, rapeseed and
soybeans to biodiesel. The program will demand about 1.4% of countrys agricul-
ture land. The country will only adopt mandatory blends when domestic biofuels
production is ensured (Department of Minerals and Energy 2007).
In 2009, Mozambique Council of Ministers approved the National Biofuels

Policy and Strategy. According to the Minister of Energy, the use of E10 and
B3 would started in 2012, and the countrys productive capacity to meet domes-
tic demand, but the interest of the government is to continue to promote ethanol
production aimed at export in the coming years (Gil 2011). Kenyan biofuel pol-
icy aims to reduce oil imports by 25% by 2030 and to increase access to energy
through sustainable biofuels production. The 2009 draft of Biofuels Strategy
stress a general capacity to produce ethanol from molasses to supply E10 blends.
In 2010, the Kenyan biofuel policy strategy defined an E10 blending mandate
(equivalent to 93million liters). The national installed capacity of 125,000liters
per day is producing only 60,000litres per day due to the limited current supplies
of molasses (GTZ and Ministry of Agriculture Kenya 2008).
In Tanzania, a country that presents suitable climate conditions and available
arable land and water, the Biofuels Guidelines of December 2009, addresses key
issues related to: institutional framework; application procedure for investors;
land acquisition and use; contract farming; sustainability of bioenergy develop-
ment; avoidance of food versus fuel conflicts and sufficient value creation for
the local rural population. In Zambia, energy security and matrix diversification
are the main drivers of biofuels introduction. The Sixth National Development
Plan defines biofuel blending ratios for bioethanol and biodiesel, for the current
period up to 2015: up to E10 and B5. Ethiopia has three state owned sugar fac-
tories which have been operational for long time. The country introduced E5 in
2009, then increasing to E10 early 2012. Government says that the countrys etha-
nol blending policy has saved the country $20.5million in fuel imports since the
policy began in 2008 (Biofuels Digest 2012).
1.3.2Latin America and Caribbean
In the Latin America and Caribbean there are a growing number of nations adopt-
ing biofuels. In Argentina, national legislation defines the blend of 5% ethanol to
gasoline and the same percentage of biodiesel to diesel oil. However, regarding
ethanol, the mandate must be introduced progressively due the lack of domestic
production capacity to attend the demand (Fundacin Bariloche 2011). Uruguay
has approved Law 18,195/2007 to introduce gradually introduce ethanol blends up
to a mixture of 5% in 2015. Domestic production is not sufficient nowadays to
supply the internal market, though a project being developed by the state-owned
oil company, is promoting the development of sugarcane crops, especially in least
developed regions of the country (Fundacin Bariloche 2011). Paraguay defined
by Law 2,748/2005, from the year of 2006, the blend of ethanol to gasoline rang-
ing from 20% up to 24%. In May 2008, Decree 12,240/2008 determined the
reduction of taxes to biodiesel and ethanol e cut out importing taxes on flexible
fuels vehicles and E85 new and used vehicles (USDA 2009). In Colombia, the
promotion of biofuels initiated in 2001, due Law 693/2001 that stipulates rules to
ethanol use and determines incentives to production, trade and consumption of this
fuel (CENBIO and CENTROCLIMA 2011).
In Mexico, besides the Bioenergy Promotion and Development Law, which
aims to promote the diversification of energy matrix, the use of biofuels is not
mandatory (BIOTOP 2009). In Costa Rica, the use of ethanol was initiated in the
1970s aiming at reduce the countrys oil dependence, however it had not succeed
at that time (Nogueira 2004). In January 2008, the Biofuels National Program
was launched aiming the increase of countrys energy security and greenhouses
emissions reduction through the blend of 7.5% ethanol to the gasoline and 5% of
biodiesel to the diesel in that year, progressively increasing to E10 and B20 blends
in 2010. The lack of infrastructure obligated government to postpone the goals
(Aguero 2011), however ethanol is available in few regions in the north of the
country yet (Villegas e Campos 2011) and apparently government is not interested
in expanding ethanol supply. Other countries in the region, such as El Salvador,
Honduras, Nicaragua, Guatemala, Peru, Ecuador and the Dominican Republic
have legislation stimulating the production and use of biofuels, however there are
no mandatory blends and local production is low, absent or devoted to foreign
markets (especially European Union and United States under advantageous trade
agreements).
1.4Main Obstacles to the Development of Ethanol Industry
To analyze the production of biofuels two main aspects must be considered, the
feedstock production (agricultural side) and its processing into biofuel (industrial
side). The current yield of agriculture production in Africa is low due to the lack
of adequate agricultural management, derived from the lack of access to fertilizers,
seeds, water and training. One of the causes is the lack of access to credit for small
holders, that prevents then to buy agriculture supplies such as fertilizers, seeds and
equipment hence reducing production costs (Mitchell 2011).
Other aspects contributing to slow down development are related to the risks
to investors and include absentee or weak land tenure policies (Cotula et al. 2004)
precarious infrastructure for distribution of the final product (Cornland et al.
2001); lack of policies to guarantee the existence of a consumer market (e.g. man-
datory blending). Regarding the industrial production there are many technologies
to produce ethanol from various feedstocks. The so-called first generation biofu-
els technologies based on sucrose fermentation are well known and largely used
worldwide. The implementation of such technologies requires few adjustments to
local operation condition and the main barrier in this case would be the lack of
trained personnel.
An additional benefit from the production of biofuels is the possibility of using
crops residues, e.g. sugarcane bagasse, as fuel in cogeneration systems. Those
units can supply the energy needs of the biofuel facility and even produce surplus
electricity to feed the surrounding areas, with the advantage that the feedstock is
readily available at low (or no) cost. It can be advantageous in African countries
where access to energy services is limited, especially in rural areas. Cogeneration
technology is commercially available in a wide range of power capacities. Low
technical skills and the lack of a supportive policy and regulatory framework
prevent investments in more efficient production. The absence of attractive and
pre-determined tariffs are the major barriers to the development of a to sell the
exceeding production (Cogen for Africa n.d.).
1.5Policy Proposals
The development of a stable institutional and political environmental is required

in order to attract companies aiming to invest in biofuels production in such mar-
kets. To establish a captive market for biofuels through the adoption of manda-
tory blends is essential, however, in many countries the domestic market can be
so small, due to economic conditions and a reduce vehicles fleet, that looking at
exporting could present an interesting opportunity. In this case, fuel quality stand-
ards have to be carefully considered. A policy of prices that enables ethanol to
compete with both gasoline and sugar prices is essential, especially in the initial
stages of its introduction into the market. Also, an extensive network for distribu-
tion and retail has to be designed in order to easily offer the product to consumers.
Considering the utilization of third parties, such as small farmers, investments
in training and equipment tend to be necessary; nevertheless many of these farm-
ers do not have access to credit to invest in its production. This problem could
be partially amended through the development of funding policies and tools, such
as microcredit and rural credit. There are countries where land tenure rights are
unclear, thus the introduction and improvement of land use and ownership legal
framework is required in order to protect and guarantee small farmers rights.
Regarding environmental aspects, the development of a zoning that defines
edible areas to grow sugarcane crops and those areas that must be protected. In
the case of other environmental issues related to the use of fertilizers, pesticides
and atmospheric emissions, the use of commercial available technologies can be
adopted as best practices. Water use tends to be a more delicate topic, especially in
African countries. The reuse of the vinasse (an ethanol production by-product) for
fertirrigation can reduce the need of water withdraw.
A fundamental aspect is related to labor conditions. It is well know that usu-
ally sugarcane cutters, that represent the largest amount of workforce in this sec-
tor, have low education levels, thus continuous training programs are essential to
increase productivity and avoid accidents. Also, the endorsement of international
working treaties and the correct enforcement must avoid child labor and inappro-
priate and degrading working conditions.
The promotion of electricity surplus production through cogeneration systems
based on sugarcane bagasse still needs the creation of institutional and regula-
tory policies aiming to minimize market risks for investors. Mechanisms such the
definition of standard power purchase agreements and feed-in tariffs can create
confidence in the market, and stimulate investments in modern and efficient bio-
mass cogeneration projects.
Acknowledgments This work was partially conducted while the author was a Giorgio Ruffolo
Fellow in the Sustainability Science Program at Harvard University. Support from Italys
Ministry for Environment, Land and Sea is gratefully acknowledged.
References
Advanced Ethanol Council (AEC) (2012) Cellulosic biofuels industry progress report 2012/2013.
http://ethanolrfa.3cdn.net/d9d44cd750f32071c6_h2m6vaik3.pdf, 2012
Aguero M (2011) Recope descarta plan de etanol a corto plazo. La Nacin. 21 de abril de 2011.
www.nacion.com/2010-07-14/ElPais/NotasSecundarias/ElPais2446754.aspx (acesso em 21
de abril de 2011)
Biofuels Digest (2012) Ethiopias ethanol blending saves $20.5 million in three years. 9 de 7 de
2012. http://www.biofuelsdigest.com/bdigest/2012/07/09/ethiopias-ethanol-blending-saves-2
0-5-million-in-three-years/ (acesso em 5 de 10 de 2012)
BIOTOP (2009) Biofuel policies and legislation in Latin America. Relatrio de projeto, 42
CENBIO and CENTROCLIMA (2011) Bioenergy, rural development and poverty alleviation in
Brazil and Colombia. So Paulo, Brazil
Cogen for Africa. Key drivers and barriers for biomass cogeneration development. s.d
Cornland DW et al (2001) Sugarcane resources for sustainabledevelopment: a case study in
Luena, Zambia. Stockholm Environment Institute, Stockholm
Cotula L, Toulmin C, Hesse C (2004) Land tenure and administration in Africa: lessons of expe-
rience and emerging issues. IIED, Londres
Department of Minerals and Energy (2007) Biofuels industrial strategy of the Republic of South
Africa
European Expert Group on Future Transport Fuels (2011) Future transport fuels. European expert
group on future transport fuels
Fischer G, Hizsnyik E, Prieler S, Shah M, van Velthuizen H (2009) Biofuels and food security.
The OPEC International, Vienna
Fundacin Bariloche (2011) Bioenergy for rural development and poverty alleviationGNESD.
San Carlos de Bariloche
Gil F (2011) Biocombustveis e petrleo: Mistura obrigatria a partir de 2012. Moambique para
todos. 14 de agosto de 2011. http://macua.blogs.com/moambique_para_todos/2011/08/bioco
mbust%C3%ADveis-e-petroleo-mistura-obrigatoria-a-partir-de-2012.html (acesso em 15 de
novembro de 2011)
Goldemberg J (2007) Ethanol for a sustainable energy future. Science 315:808810
Goldemberg J (2008) The Brazilian biofuels industry. Biotechnol Biofuels, 1 de May de 2008:
16
GTZ, Ministry of Agriculture Kenya (2008) A roadmap for biofuels in Kenya
IPCC (2007) Fourth assessment report: climate change 2007synthesis report. Contribution of
Working Groups I, II and III to the Fourth Assessment Report of the Intergovernmental Panel
on Climate Change. IPCC, Geneva
Janssen R, Rutz D (2012) Bioenergy for sustainable development in Africa. Springer, Berlin
Mitchell D (2011) Biofuels in Africa: opportunities, prospects, and challenges. The World Bank,
Washington, DC
Nogueira LAH (2004) Perspectivas de un programa de biocombustibles en Amrica. Cidade do
Mxico: CEPAL
OECD/IEA (2010) World energy outlook 2010. International Energy Agency, Paris
REN21 (2012) Renewables 2012 global status report. REN21 Secretariat, Paris
RFA (2011) World fuel ethanol production 2011. http://ethanolrfa.org/pages/World-Fuel-Ethanol-
Production (acesso em 15 de novembro de 2011)
Smeets EMW, Andr PC, Faaij APC, Lewandowski IM, Turkenburg WC (2007) A bottom-up
assessment and review of global bio-energy potentials to 2050. Prog Energy Combust Sci
33:56106
USDA (2009) Paraguay biofuels annual. Buenos Aires
Villegas JS, Campos C (2011) Gasolina con alcohol seduce a choferes de Guanacaste. La
Nacin. 21 de 04 de 2011. http://www.nacion.com/2011-04-19/ElPais/Relacionados/ElPais2
753442.aspx (acesso em 25 de maio de 2011)
Worldwatch Institute (2007) Biofuels for transport. Springer, London
Chapter 2
Can Energy Policy Drive Sustainable
Land Use? Lessons from Biofuels Policy
Development Over the Last Decade
Jeremy Woods and Nicole Kalas
2.1Introduction
The mandated increase in bioenergy as a means to decarbonise our energy supply,

enhance energy security, and promote rural development has raised concerns regard-
ing the impacts biomass feedstock production may have on food security. These
national mandates appear to have placed bioenergy feedstock production in competi-
tion for resources required to feed a growing global population. In turn, concerns over
the direct and indirect impacts of bioenergy, particularly conventional biofuels,1 have
pushed policy makers to try to direct biomass crop production for energy onto mar-
ginal, degraded and unused land. Moving bioenergy onto marginal lands will inevi-
tably raise the costs of feedstock production, but it may also be contradictory to food
security where sustainable intensification and reduced losses require increased
energyinputs into agriculture. This marginalisation ignores the beneficial role that
perennial energy crops could play in managing the sustainable intensification of over-
all agricultural production required to feed over 9 billion people by 2050. Chapter 2,
therefore, explores the role and drivers of bioenergy in future world energy produc-
tion, land use change and wider sustainability issues, and proposes an alternative,
integrated approach toward a resource efficient and sustainable provision of agricul-
tural products, including food, feed, biobased chemicals, materials and energy.
1 Conventional biofuels are produced through the fermentation of sugars or starches to bioetha-
nol from commodity crops such as sugarcane, maize, wheat, and beet, or through the methyl
esterification of vegetable oils to biodiesel from palm, soy or oilseed rape.
J. Woods(*) N. Kalas
Centre for Environmental Policy, Centre for Energy Policy and Technology,
Imperial College London, London, UK
e-mail: jeremy.woods@imperial.ac.uk
N. Kalas
e-mail: n.kalas@imperial.ac.uk

14 J. Woods and N. Kalas
2.1.1Changing Patterns in Global Energy Supplies
Two concurrent developments in energy use are changing the pattern of global
energy supplies (see Fig.2.1). On one side, climate policy and energy security
driven increases in efficiency and uptake of renewable energy technologies in
the USA, European Union and Japan are slowly reversing the upward trend of
oil imports observed to date. In the USA, increased domestic production of oil
and the recent intensification in shale gas exploration (hydraulic fracturing) have
placed the country on a path to energy self-sufficiency. On the other side, rapid
economic growth in China and India, driven primarily by fossil fuels, has led to
increased consumption of cheap coal and imported oil. These opposing trends in
oil consumption are raising the competition for energy security and shifting the
global balance of oil imports from OECD to non-OECD countries, where China is
expected to become the worlds largest oil importer by 2020.
As Fig.2.2 illustrates, new coal has provided nearly 50% of incremental
energy supply since 2000 and in increasingly inefficient power plants (to lower
capital costs). In the USA, shale gas has started to drive coal out of the electric-
ity generation mix and is also degrading the role for dedicated biomass and other
sources of renewable energy. In the UK, power generators are moving rapidly
towards large scale biomass co-firing in existing electricity plants and reducing
demand for dedicated biomass.
Cheap coal and the shale gas revolution are the biggest challenges to climate
change mitigation and the meaningful deployment of renewables. The world is
not on track to meet the internationally agreed target to limit the long-term rise in
the global average temperature to 2C. Over 80% of global energy consumption
is based on fossil fuels, and the energy sector accounts for approximately 2/3 of
Greenhouse Gas (GHG) emissions (IEA 2013).
Fig.2.1Net imports of oil (20002035). Source IEA (2011)

2 Can Energy Policy Drive Sustainable Land Use? 15
Fig.2.2Growth in global
energy demand (20002010).
Source IEA (2011)
2.1.2Future World Energy Production and Price Trends:

Is Bioenergy Policy Swimming Against the Tide?
2.1.2.1World Energy Production and Price Trends
The IEAs Current Policies Scenario2 shows an increase in world oil production by
26% from 82.3mb/d3 (2007) to 103.8mb/d in 2030 (see Fig.2.3) (IEA 2008). To
meet demand growth and offset decline, an additional 64mb/d would be needed,
which corresponds to six times Saudi Arabias current capacity.
Figure2.4 shows the global trends in prices for fossil fuels ($/GJ) indicating a
continued increase in oil prices, and recent decrease in both gas and coal (BP 2012).
Gas and coal are expected to resume their upward trend, but stay below oil prices.
2.1.2.2Global Bioenergy Policy and Consumption
Bioenergy policies are motivated by climate change mitigation targets, energy

access and security, and rural development. While global demand for biomass
feedstocks is predominantly driven by policies in the EU and USA, at least 33
countries have now implemented mandates for biofuels (blending requirements)
(Biofuel Digest 2012).
In the EUs Renewable Energy Directive (RED), Member States have commit-
ted to reduce their CO2 emissions by 20% and to target a 20% share of renewable
energies in the EU energy mix (including 10% of transport fuels by 2020 as part
of the 2007 The EU climate and energy package (EC 2009a)). Biofuel demand is
projected to be 7,307ktoe4 (14,450 million litres) of ethanol, and 21,650ktoe
2The IEAs Current Policies Scenario (previously called the Reference Scenario) assumes no
changes in energy and GHG emission reduction policies (IEA 2010).
3mb/d=million barrels/day (1 barrel=159litres).
4Ktoe=kilo tonnes of oil equivalent (1ktoe EtOH=1.978million litres, 1ktoe bio-
diesel=1.32 million litres).

Fig.2.3World oil production by source in the Current Policies Scenario (19902030). Source
IEA (2008)
Fig.2.4Global trends in fossil fuel prices (19702011). Source BP (2012)
(28,600 million litres) of biodiesel (Beurskens et al. 2011). The dominant driver
for the RED is GHG mitigation, but energy security is also a serious concern. In
conjunction with the Fuel Quality Directive (FQD), sustainability criteria for GHG
emission reductions and biodiversity conversion are applied to each supply chain
through assurance and certification schemes (EC 2009a, b, 2012).
In the USA, biofuel blending is mandated by the Renewable Fuels Standard
(RFS2) to achieve the targets established in The Energy Independence and Security
Act of 2007 (EISA). The RFS2 has laid the foundation for achieving significant
GHG emissions in the transport sector and for promoting the development of the
US renewable fuels sector. It provides volumetric standards for renewable fuels,
including advanced biofuels, and includes GHG emission thresholds producers are
Fig.2.5World bioenergy use by sector and use of traditional biomass in the IEA new policies sce-
nario (2010 and 2035). Source IEA (2012)
required to meet. Under the RFS2, annual biofuel production, which in the USA is
predominantly ethanol, is to increase from currently 13.2billion gallons (60billion
litres) (2012) to 36billion gallons (164billion litres) by 2022 (EPA 2010).
In 2010, total global bioenergy consumption amounted to 1,277Mtoe,5 or exclud-
ing traditional biomass,6 526Mtoe (see Fig.2.5). The IEA New Policies Scenario7
estimates that in 2035 total world bioenergy use will increase to 1,881Mtoe, or
1,200Mtoe excluding traditional biomass (at an average annual growth rate of 3.3%).
Currently, the industrial sector is the largest consumer of bioenergy (196Mtoe), but
the power sector will dominate bioenergy consumption in 2035 (414Mtoe). Together,
the power and industrial sector will demand approximately 2/3 of global bioenergy in
2035. The use of traditional biomass will continue to decline, as access to modern and
more efficient energy technologies, including modern bioenergy,8 increases in devel-
oping countries. Excluding the use of traditional biomass, the EU will be the single
largest consumer of bioenergy, increasing its consumption from 130Mtoe (2010) to
230Mtoe (2035), whereas the US will follow closely with 210Mtoe by 2035 (IEA
2012). Global biofuel (or liquid bioenergy) consumption, dominated by ethanol, is
estimated to increase by 250% to 210Mtoe during that period, driven primarily by
blending mandates (IEA 2012).
5Mtoe=million tonnes of oil equivalent (1Mtoe=41.9PJ).

6Traditional biomass includes wood, charcoal, crop residues and animal dung and is mainly
used for heating and cooking (IPCC 2011).
7 The New Policies scenario is IEAs central scenario and takes into account the cautious imple-
mentation of broad policy commitments and plans to address energy and GHG emission reduc-
tion challenges (IEA 2012).
8 Modern bioenergy is utilised at higher efficiencies than traditional biomass and includes liq-
uids and gases as secondary energy carriers to generate heat, electricity, combined heat and
power (CHP), and transport fuels (IPCC 2011).
2.2What is Sustainable Bioenergy and What to Measure
The main areas of concern for policymakers regarding the sustainability of bioen-
ergy production (in particular that of biofuels) are its impacts on food security and
global commodity prices, life cycle GHG emissions reductions, resources deple-
tion, land grabbing, ecosystem services and biodiversity. Figure2.6 shows that
sustainability of bioenergy needs to be considered systematically and holistically
across the three pillars of sustainability (environmental, social and economic). It
also points out the importance of scale and geographic context in the sustainability
assessment of bioenergy value chains.
The EU, which depends more on imported feedstocks than the USA, both in
terms of amounts and variety, to meet its bioenergy demands has been on the fore-
front of formulating broad environmental sustainability safeguards into its regula-
tions (FAO 2013). However, the implementation of these criteria is complicated by
the fact that many feedstocks have multiple, substitutable end-uses, e.g., wheat is
used for food, feed, and fuel production, whereas the criteria apply to a single end-
use thus creating the potential for leakage (Frank et al. 2012). Furthermore, at pre-
sent, social sustainability safeguards are only realised as part of voluntary schemes
adopted by selected biofuels producers.
The sustainability of bioenergy in terms of their efficacy to reduce GHG
emissions by substituting fossil fuels hinges on two main factors: land use and
biomass production practices. Land use change has direct (positive or negative)
implications on terrestrial carbon stocks, and management practices encompass-
ing zoning, crop selection and cultivation, energy and fertiliser inputs impact the
GHG balance of the end-product. The core of the debate about the efficacy of
bioenergy (again, with a particular focus on biofuels) continues to centre on the
issue of indirect land use change (ILUC). While some modelling results indicate
no ILUC impacts (e.g., Kim and Dale 2011), other studies show significantly
lower impacts than previously estimated (e.g., INRA 2013) or very high GHG
Fig.2.6Measures of sustainable bioenergy

Fig.2.7Uncertainties of estimated indirect land use change GHG emission for selected biofuels
(g CO2eq/MJ). Source adapted from EC DG-Tren (2010)
emissions (e.g., Searchinger 2010). Figure2.7 further illustrates the divergence

in the results of different ILUC modelling studies. The debate around the signifi-
cant scientific uncertainties as to the magnitude and effect of ILUC has slowed
down the development of bioenergy supply chains and diverted attention from
wider issues of the sustainability of bioenergy and agricultural production more
broadly.
Land grabbing, defined as the transfer of the right to own or use the land
from local communities to foreign investors through large-scale land acquisitions
(Rulli et al. 2012) has also been attributed to the increase in demand for bioenergy
feedstocks (GRAIN 2013). While numerous cases of illegal appropriations and
human rights violations with disastrous impacts on smallholders and local com-
munities have been reported and must be prevented in the future, a recent analy-
sis by the Land Matrix (2013) suggests that the scale of the problem may have
been largely exaggerated. Land Matrix reviewed 950large-scale land acquisitions
(LSLA) of 200ha or more since 2000. Of the 750concluded deals, covering a
total area of 32.6Mha, their research concludes that only approximately 5% (or
1.63Mha) have gone into agricultural production. Figure2.8 shows that that while
biofuel production has had an impact, food crops accounted for a larger share of
deals and area. Forestry and tourism were also important sources of demand for
land. A study by IIED on the socio-economic impacts of such land acquisitions
concludes that the impact of these investments depends on the way they are struc-
tured, and can either create new opportunities to improve local living standards,
or further marginalise the poor (IIED 2009).
Fig.2.8Main drivers of large-scale land acquisitions. Source Land Matrix (2013)
Biofuels have also been blamed for the 2008-2009 spikes in food prices (e.g.,
Pimentel et al. 2009; ActionAid 2010). However, recent studies indicate that the
causal relationships are more complex and that the increase in commodity prices
can be primarily attributed to high crude oil prices (affecting energy and fertiliser
costs), exchange rate movements, stock-to-use ratios, unusually frequent adverse
weather events, and only in small part to EU and US demands for conventional bio-
fuel feedstocks (Baffes and Dennis 2013; Oladosu and Msangi 2013). Nevertheless,
concerns over global food security have dichotomised the issue and effectively
placed the production of food and fuel in opposition (Rosillo-Calle 2012).
Sustainable bioenergy production must also adequately consider the protection
of biodiversity. According to the Millennium Ecosystem Assessment (2005), see
Fig.2.9, current rates of species extinction are at least two orders of magnitude above
background rates and are expected to rise to at least three orders above background
rates. In the UK, 60% of monitored species have declined over the past 50years
and 10% of species are threatened by extinction (UK 2013). Drivers of this unprec-
edented rate of biodiversity loss are habitat conversion and fragmentation, primarily
due to agricultural expansion and urban development; increasingly, climate change,
which contributes to habitat change, is becoming the dominant driver of extinction.
To address the aforementioned concerns regarding the sustainability of bioen-
ergy and to provide policymakers and producers with a comprehensive framework
to promote and monitor the development of bioenergy supply chains, the Global
Bioenergy Patnership (GBEP) Proposed 24 indicators for the sustainable produc-
tion and use of modern bioenergy (GBEP 2011). Table2.1 summarises these indi-
cators by pillars and themes. These indicators are thus far the only comprehensive
framework for the sustainable development of bioenergy.
Fig.2.9Species extinction (per thousand species per millennium). Source MEA (2005)
2.2.1Sustainable Agricultural Intensification: The Future

of Food and Farming: Five Challenges for Global
Sustainability
The production of bioenergy sits within a larger system of agricultural produc-

tion. Bioenergy policies, given their narrow scope and mandate, cannot address the
inefficiencies of global agricultural production overall. However, the controversies
surrounding the large scale deployment of bioenergy, such land use change, food
versus fuel, land grabbing, biodiversity loss, etc. may have assisted in recognis-
ing the necessity for a profound shift from conventional agricultural practices to a
more sustainable, resource efficient and climate-smart, multi-product agricultural
production system.
The need to provide food, shelter, energy and other resources for 9.2 billion
people in 2050 against the backdrop of climate change requires concerted efforts
today to avoid future shocks to global food production (Foresight 2011; Garnett
et al. 2013). The Future of Food and Farming report highlights five key challenges
for global food system (Foresight 2011):
A. Balancing future demand and supply sustainablyto ensure that food supplies
are affordable.
Table2.1GBEP sustainability indicators
22
Pillars
GBEPs work on sustainability indicators was developed under the following three pillars, noting interlinkages between them:
Environmental Social Economic
Themes
GBEP considers the following themes relevant, and these guided the development of indicators under these pillars:
Greenhouse gas emissions, productive capacity of Price and supply of a national food basket, Resource availability and use efficiencies in
the land and ecosystems, air quality, access to lard, water and other natural bioenergy production, conversion, distribution
water availability, use efficiency and quality, resources, labour conditions, rural and and end use, economic development, economic
biological diversity, land-use change, social development, access to energy, viability and competitiveness of bioenergy,
including indirect effects human health and safety access to technology and technological capabili-
ties, energy security/diversification of sources
and supply, energy security/Infrastructure and
logistics for distribution and use
Indicators
1. Lifecycle GHG emissions 9. Allocation and tenure of land for 17. Productivity
new bioenergy production
2. Soil quality 10. Price and supply of a rational food basket 18. Net energy balance
3. Harvest levels of wood resources 11. Change in income 19. Gross value added
4. Emissions of non-GHG air pollutants, 12. Jobs in the bioenergy sector 20. Change in consumption of fossil fuels
including air toxics ard traditional use of biomass
5. Water use and efficiency 13. Change in unpaid time spent by 21. Training ard regualification of the workforce
women and children collecting biomass
6. Water quality 14. Bioenergy used to expand access to modern 22. Energy diversity
energy services
7. Biological diversity in the landscape 15. Change in mortality and burden of 23. Infrastructure and logistics for distribution of
disease attributable to indoor smoke bioenergy
8. Land use and land-use change related to 16. Incidence of occupational injury, 24. Capacity and flexibility of use of bioenergy
bioenergy feedstock production illness and fatalities
Source GBEP (2011)
J. Woods and N. Kalas
B. Ensuring that there is adequate stability in food suppliesand protecting the

most vulnerable from the volatility that does occur.
C. Achieving global access to food and ending hungerfood security for all.
D. Managing the contribution of the food system to the mitigation of climate
change.
E. Maintaining biodiversity and ecosystem services while feeding the world.
Because of the size of future threats and increasing demands that go beyond
food production, such as climate change mitigation and ecosystem service provi-
sion, a radical redesign is needed. No action is not an optionif the food system
fails to deliver against these future challenges the implications will be profound,
and, aside from the human tragedy, threaten political stability and security.
2.3Bioenergy and Land Use
2.3.1Technical Bioenergy Potentials and Global Land Availability
Figure 2.10 is based on IEA (2009) data and shows the relative contribution of
global bioenergy (including electricity, district and onsite heat, and biofuels) to
global energy demand, assuming a 1.5% CAGR9 for conventional energy, a 5%
CAGR for modern bioenergy, and 1.2% for traditional biomass. In this scenario,
bioenergy provision would equal approximately 250 EJ, the equivalent to roughly
25% of global primary energy demand, if business as usual trends continue.
Based on these results, it was calculated that in 2050 (beyond the 2035 IEA sce-
narios), approximately 100650Mha (16.5million km2) of land would be
required for the production of the necessary biomass (Murphy et al. 2011).
The range corresponds to similar, lower range estimates found in the literature
(see Table2.2), and can be explained in part by the inherent uncertainty about
land availability and productive potentials in the future. However, a reasonable
share of the range in published potentials can also be explained by the individual
assessments differing focus on fundamental theoretical potentials versus real-
istic exportation potentials. When evaluating the bioenergy resource and poten-
tial for exploiting biomass for bioenergy at national to global scales, Sathaye (in
Brown et al. 1996) outlines the following five step progression from theoretical to
practical/realisable potentials:
1. Biological/theoretical potential
2. Technological potential
3. Economic potential
4. Ecological potential
9 CAGR: Compounded annual growth rate.

Fig.2.10Global bioenergy versus primary energy demand (20062050). Source calculations

based on IEA (2009)
5. Realistic potential/implementation.
Furthermore, as Fig. 2.11 shows, land availability is unevenly distributed
among different regions. Latin America and the Caribbean and Sub-Saharan Africa
are the only two regions where substantial amounts of suitable land may still be
available and where agricultural yields could be significantly increased through
improved inputs and management practices.
2.3.2Energy Crop Yield Estimates
The productivity of agricultural crops depends on climate, soil conditions, and

agricultural management practices, including types of cultivars, the quality of
seeds, availability of water, agrochemical inputs, and pest and control. Given the
competing demands for land to meet of human needs for (1) supply of resources,
(2) provision of ecosystem services, and (3) space for human infrastructure
(Dunlap and Catton 2002), further expansion of cropland is limited in the longer
term. Instead, future yield increases will have to be achieved through higher per
hectare productivity. In part, this can be attained by closing the yield gap between
attainable and observed yields across regions (Mueller et al. 2012). Haberl et al.
(2007) estimate that global productivity of cropland is currently 35% below its
potential productivity. Additional gains will come from a more efficient utilisa-
tion of all harvested plant parts in integrated production systems to meet human
demand for food, feed, fibre, chemicals, and energy.
Figure 2.12 maps the results of multiple studies on energy crop potentials
against available land areas. Generally, the data points indicating yields below
5odt10/ha assume production on marginal and degraded lands, while those above
10odt=oven dry tonne.

Table2.2Overview of recent studies on technical potentials of biomass from energy crops
Reference Type of Regions Time Sustainability Land use types Land area Productivity Potential of
potential frame constraints used [mio. [tonnes dry energy crops
Km2] matter/ha/yr] [EJ/yr]
van Technical Global 2050 Biodiversity, food Abandoned agricultural 13 Depending on 120-300 EJ/
Vuuren security, soil land (75%) Grassland land suit- yr (uncon-
et al. degradation, water (25%) ability and strained)
(2009) scarcity climate factors 65-115 EJ/yr
1.0-3.2kg dry (constrained)
matter/m3/yr
WBGU Technical Global 2050 Biodiversity, C bal- Land suitable for bioenergy 2.45.0 7.5-12.6 t/ha/yr 34-120 EJ/yr
(2008) ance, deforestation, cultivation according
degraded land, to the crop functional
food security, types in the model, con-
water scarcity sidering sustainability
Campbell Technical Global 2000 (not Agricultural lands, Abandoned agricultural 3.94.7 4.3t/ha/y (AGB) 3241EJ/yr
et al. clearly ecosystems, food land (100%) (AGB)
2 Can Energy Policy Drive Sustainable Land Use?
(2008) men- security, releas-

tioned) ing carbon stored
in forests, water
scarcity
Field et al. Technical Global 2050 Biodiversity, food Abandoned agricultural 3.9 3.2tC/ha/yr 27EJ/yr (AGB)
(2008) security. ecosys- land (100%)
tems, deforestation
Dornburg Technical Global 2050 Land for food Not explicitly specified Not specified Not specified Energy crops:
et al. excluded various 120EJ/yr
(2010) assumptions on
(non-) exclusion
of degraded and
protected land
(continued)
25
Table2.2continued
26
Reference Type of Regions Time Sustainability Land use types Land area Productivity Potential of
potential frame constraints used [mio. [tonnes dry energy crops
Km2] matter/ha/yr] [EJ/yr]
Smeets Technical 11 world 2050 Biodiversity, defor- Surplus agricultural land 7.335.9 1621 odt(oven 2151,272EJ/yr
et al. regions estation, food (100%) dry tonnes)/
(2007) security ha/yr
Hoogwijk Technical 11 world 2050-2100 Biodiversity, food Abandoned agricultural Abandoned: Depending on Abandoned:
et al. regions security land (100%) remain- 0.61.5 land suitabil- 130400EJ/
(2005) ing land not for food rest land ity and climate yr rest land
or material procution 0.31.4 factors 235240EJ/
(1050%) extensive yr total:
grassland 300650EJ/
yr
Erb et al. Technical 11 world 2050 Excluded: land for Cropland not needed for 2.39.9 Equal to potential Bioenergy
(2009) regions food and feed, food and fiber supply depend- (cropland) or crops:
forestry and unpro- intensification of grazing on actual (graz- 28128EJ/
ductive land ing land food ing land) NPP yr residues:
and feed 2136EJ/yr
demand
(44
scenarios)
Source IIASA (2012)
J. Woods and N. Kalas
Fig.2.11Global land use and availability. Source PCFISU (2011based on IIASA GAEZ
study)
15odt/ha assume production on good quality land and high yielding crop varieties
(Slade et al. 2011). In some regions, e.g., Brazil, bioenergy crop yields (sugarcane)
already exceed 18odt/ha.
2.3.3Boxing in Bioenergy
Concerns over the direct and indirect impacts of bioenergy, particularly conven-
tional biofuels, have pushed policy makers to try to direct biomass crop produc-
tion to marginal, degraded and unused land. This will not only raise the costs of
feedstock production and transport, but may also be contradictory to food secu-
rity where sustainable intensification and reduced losses require increased energy
inputs. Furthermore, this marginalisation ignores the beneficial role that perennial
energy crops could play in integrated production systems by managing intensifica-
tion through nutrient capture and water quality benefits, watershed, soil and ero-
sion protection, and increased biodiversity.
2.3.4Integrating Biomass Supply Chains and Sustainable

Biorenewables Innovation
To transcend the inefficiencies and inadequacies of current bioenergy supply

chains, bioenergy production needs to be embedded within a broader, integrated
system of biomass productions systems which are optimised for the sustainable
Fig.2.12Energy crop yield estimates. Source adapted from Slade et al. (2011)
provision of food, feed, fuel, fibres, chemicals, energy and ecosystem services. The
integration of biomass supply chains aims to optimise the use of resources in agri-
cultural supply chains and exploit its maximum value along each step, as illustrated
in Fig.2.13. Initial studies indicate that integrated food and energy systemsor
multi-functional production systemscan increase overall yields per unit of land
area, provide important socio-economic benefits through the diversification on
marketable products for farmers, and have positive impacts on the environment,
including enhanced carbon sequestration (FAO 2010; Bogdanski 2012; Dale et al.
2010). Research is underway on how the inclusion of perennial buffer strips can
trap nutrient run-off along arable land adjacent to rivers and streams while provid-
ing valuable feedstocks for biobased products and advanced biofuels, or using per-
ennials for the restoration of degraded land (Gopalakrishnan et al. 2009, 2012).
Novel approaches to integrated feedstock production are only the first step in sus-
tainable biomass supply chains and will only be able make a difference at scale if
integrated with efficient downstream processing and access to markets. Figure2.14
shows the main technological conversion pathways and inter-linkages for biomass
feedstocks which enable an optimised, or cascading (Haberl and Geissler 2000),
use of all plant components, including the utilisation co-products (Black et al. 2011).
2.4How Can Bioenergy Policy Drive Sustainable

Land Use?
Bioenergy policies were designed to promote the use of bioenergy as a means to

reduce GHG emissions, increase energy access and security and stimulate rural devel-
opment. Overall, sustainability considerations, with the exception of GHG emission
Fig.2.13Integrating biomass supply chains
Fig. 2.14Feedstock and technology pathways for biorenewablesmany options. Source

Adapted from Black et al. (2011)
reduction requirements and some provisions for the protection of high biodiversity
areas, have not been adequately addressed to date. Managing sustainability effectively
is challenging, and needs to balance all three of its pillarsand possible trade-offs
between themacross different spatial and temporal scales and management systems.
In contrast to other renewable energy technologies, bioenergy is inextricably
linked to land use, agricultural production and forestry. The attempt to regulate
bioenergy feedstock production without addressing the sustainability of agricul-
tural production overall, will continue to result in leakage and do little to resolve
the deadlocked food versus fuel versus biodiversity debate.
Current agricultural land use practices are not sustainable and pushing future
energy crop production onto marginal lands, as some scenarios envision, may in
the long run increase the economic, social, and environmental costs of bioenergy
production. Furthermore, it would miss the opportunity to use bioenergy markets
as a vehicle to promote the necessary sustainable intensification of agriculture in
areas that lag behind in terms of agricultural productivity, are least food secure and
suffer disproportionately from the impacts of climate change.
Instead of continuing with business as usual, this chapter offers an alternative
path forward, and shows that by integrating the production of food, feed, fuel,
fibres, chemicals, and energy the environmental, social and environmental sustain-
ability of agricultural production systems could be greatly enhanced.
Bioenergy policy alone cannot effectively drive sustainable land use. Given the
many demands on land, land use planning is inherently complex and intensive,
and past experience has shown that complex policy is often ineffective. One way
for regulators to overcome this dilemma in the area of bioenergy and land use, is
to incentivise (through a mix of regulation and market-based incentives) a shift
to more resource use efficient, integrated production systems as an integral com-
ponent in the transition to a low carbon economy, with potential linkages to pay-
ments for ecosystem services and carbon sequestration schemes.
References
ActionAid (2010) Meals per gallonthe impact of industrial biofuels on people and global
hunger. London. http://www.actionaid.org.uk/sites/default/files/doc_lib/meals_per_gallon_
final.pdf
Baffes J, Dennis A (2013) Long-term drivers of food prices. Policy research working
paper 6455. World Bank, Washington DC. http://papers.ssrn.com/sol3/papers.cfm?
abstract_id=2268316##
BP (2012) BP statistical review of world energy. June 2012. London. http://www.bp.com/content/
dam/bp/pdf/Statistical-Review-2012/statistical_review_of_world_energy_2012.pdf
Beurskens LWM, Hekkenberg M, Vethman P (2011) Renewable energy projections as published
in the national renewable energy action plans of the European member statessummary
report. ECN, EEA, Copenhagen. http://www.ecn.nl/docs/library/report/2010/e10069.pdf
Biofuel Digest (2012) Biofuels mandates around the world: 2012. http://www.biofuelsdigest.com
/bdigest/2012/11/22/biofuels-mandates-around-the-world-2012/ (accessed on 4 July 2013)
Black MJ, Whittaker C, Hosseini SA, Diaz-Chavez R, Woods J, Murphy RJ (2011) Life cycle
assessment and sustainability methodologies for assessing industrial crops, processes and end
products. Ind Crops Prod 34(2):13321339. doi:10.1016/j.indcrop.2010.12.002
Bogdanski A (2012) Integrated foodenergy systems for climate-smart agriculture. Agri Food
Secur 1:9 (http://www.agricultureandfoodsecurity.com/content/pdf/2048-7010-1-9.pdf)
Brown S, Sathaye J, Cannell M, Kauppi P, Burschel P, Grainger A, Heuveldop J, Leemans
R, Moura Costa P, Pinard, Nilsson S, Schopfhauser W, Sedjo R, Singh N, Trexler M, van
Minnen J, Weyers S (1996) Management of forests for mitigation of greenhouse gas emis-
sions. In: Watson RT, Zinyowera MC, Moss RH, Dokken DJ (eds) IPCC climate change
1995impacts, adaptations and mitigation of climate change: scientific-technical analyses,
contribution of working group II to the 2nd assessment report of the intergovernmental panel
on climate change. Cambridge University Press, Cambridge, pp 773797. https://docs.goog
le.com/uc?export=download&confirm=no_antivirus&id=0B1gFp6Ioo3akQS1GX3owSk0
tLU0
Campbell JE, Lobell DB, Genova RC, Field CB (2008) The Global Potential of Bioenergy on
Abandoned Agriculture Lands. Environ Sci Technol 42(15):57915794
Dale B, Bals BD, Kim S, Eranki P (2010) Biofuels done right: land efficient animal feeds
enable large environmental and energy benefits. Environ Sci Technol 44:83858389.
doi:10.1021/es101864b
Dornburg V, van Vuuren D, van de Ven G, Langeveld H, Meeusen M, Banse M, van Oorschot M,
Ros J, van den Born GJ, Aiking H, Londo M, Mozaffarian H, Verweij P, Lyseng E, Faaij A
(2010) Bioenergy revisited: Key factors in global potentials of bioenergy. Energy Environ Sci
3:258267
Dunlap RE, Catton WR (2002) Which function(s) of the environment do we study? A compar-
ison of environmental and natural resource sociology. Soc Natural Resour 15(3):239249.
doi:10.1080/089419202753445070
EC (2009a) Directive 2009/28/EC of the European Parliament and of the council of 23 April
2009 on the promotion of the use of energy from renewable sources and amending and
subsequently repealing Directives 2001/77/EC and 2003/30/EC. http://eur-lex.europa.
eu/LexUriServ/LexUriServ.do?uri=CELEX:32009L0028:en:NOT
EC (2009b) Directive 2009/30/EC of the European Parliament and of the council of 23 April
2009 amending Directive 98/70/EC as regards the specification of petrol, diesel and gas-oil
and introducing a mechanism to monitor and reduce greenhouse gas emissions and amending
Council Directive 1999/32/EC as regards the specification of fuel used by inland waterway
vessels and repealing Directive 93/12/EEC. http://eur-lex.europa.eu/LexUriServ/LexUriServ.
do?uri=CELEX:32009L0030:EN:NOT
EC (2012) Biofuelssustainability schemes. Recognised voluntary schemes. http://ec.europa.eu/
energy/renewables/biofuels/sustainability_schemes_en.htm (accessed on 4 July 2013)
EC DG-Tren (2010) Assessment of the impact of land use change on greenhouse gas emissions
from biofuels and bioliquids. Incomplete draft; version 4 1 10
EPA (2010) 40 CFR Part 80. Regulation of fuels and fuel additives: changes to renewable fuel
standard program, final rule, 3 Feb 2010 (RFS2). http://www.gpo.gov/fdsys/pkg/FR-2010-
03-26/pdf/2010-3851.pdf (accessed on 4 July 2013)
Erb K, Haberl H, Krausmann F, Lauk C, Plutzar C, Steinberger JK, Mller C, Bondeau A, Waha
K, Pollack G (2009) Eating the Planet: Feeding and fuelling the world sustainably, fairly and
humanelya scoping study. Institute of Social Ecology, Vienna
FAO (2010) making integrated food-energy systems work for people and climate. Rome.
http://www.fao.org/docrep/013/i2044e/i2044e.pdf
FAO (2013) Biofuels and the sustainability challenge: a global assessment of sus-
tainability issues, trends and policies for biofuels and related feedstocks. Rome.
http://www.fao.org/docrep/017/i3126e/i3126e.pdf
Field CB, Campbell JE, Lobell DB (2008) Biomass energy: the scale of the potential resource.
Trends Ecol Evol 23(2):6572
Foresight (2011) The future of food and farming. The Government Office for Science, London.
http://www.bis.gov.uk/assets/foresight/docs/food-and-farming/11-546-future-of-food-
and-farming-report.pdf
Frank S, Bttcher H, Havlk P, Valin H, Mosnier A, Obersteiner M, Schmid E, and Elbersen B
(2012) How effective are the sustainability criteria accompanying the European Union 2020
biofuel targets? GCB Bioenergy
Garnett T, Appleby M, Balmford A, Bateman I, Benton T, Bloomer P, Burlingame B, Dawkins
M, Dolan L, and Fraser D (2013) Sustainable intensification in agriculture: premises and pol-
icies. Sci 341(6141):3334
GBEP (2011) The global bioenergy partnership sustainability indicators for bioenergy. Rome.
http://www.globalbioenergy.org/fileadmin/user_upload/gbep/docs/Indicators/The_GBEP_
Sustainability_Indicators_for_Bioenergy_FINAL.pdf
Gopalakrishnan G, Negri MC, Wang M, Wu M, Snyder SW, LaFreniere L (2009) Biofuels, land,
and water: a systems approach to sustainability. Environ Sci Technol 43(15):60946100.
doi:10.1021/es900801u
Gopalakrishnan G, Negri MC, Salas W (2012) Modeling biogeochemical impacts of bioen-
ergy buffers with perennial grasses for a row-crop field in Illinois. Glob Change Bioenergy
4(6):739750. doi:10.1111/j.1757-1707.2011.01145.x
GRAIN (2013) Land grabbing for biofuels must stopEU biofuel policies are displac-
ing communities and starving the planet. Barcelona. http://www.grain.org/article/
entries/4653-land-grabbing-for-biofuels-must-stop
Haberl H, Geissler S (2000) Cascade utilization of biomass: strategies for a more efficient use of
a scarce resource. Ecol Eng 16:111121. doi:10.1016/S0925-8574(00)00059-8
Haberl H, Erb KH, Krausmann F, Gaube V, Bondeau A, Plutzar C, Gingrich S, Lucht W,
Fischer-Kowalski M (2007) Quantifying and mapping the human appropriation of net pri-
mary production in earths terrestrial ecosystems. Proc Nat Acad Sci 104(31):1294212947
(http://www.pnas.org/content/104/31/12942)
Hoogwijk M, Faaij A, Eickhout B, de Vries B, Turkenburg W (2005) Potential of biomass energy
out to 2100, for four IPCC SRES land-use scenarios. Biomass Bioenergy 29(4):225257
IEA (2008) World energy outlook 2008. Paris. http://www.worldenergyoutlook.org/
media/weowebsite/2008-1994/weo2008.pdf
IEA (2009) World energy outlook 2009. Paris. http://www.worldenergyoutlook.org/media/weowe
bsite/2009/WEO2009.pdf
IEA (2010) World energy outlook 2010. Paris. http://www.worldenergyoutlook.org/media/
weo2010.pdf
IEA (2011) World energy outlook 2011. Paris. http://www.worldenergyoutlook.org/publications/
weo-2011/
IEA (2012) World energy outlook 2012. Paris. http://www.worldenergyoutlook.org/publications/
weo-2012/
IEA (2013) World energy outlook special report. Redrawing the energyclimate map. Paris. http:/
/www.worldenergyoutlook.org/energyclimatemap/#d.en.36900
IIASA (2012) Global energy assessment (GEA). Global energy assessmenttoward a sustain-
able future. Cambridge University Press, Cambridge, UK and New York, NY, USA and the
International Institute for Applied Systems Analysis, Laxenburg, Austria. http://www.globale
nergyassessment.org/
IIED (2009) Land grabs in Africa: can the deals work for development? IIED Briefing, Sept
2009. London. http://pubs.iied.org/pdfs/17069IIED.pdf
INRA (2013) Le changement daffectation des sols induit par la consommation europenne
de biodiesel: une analyse de sensibilit aux volutions des rendements agricoles. Rennes.
http://prodinra.inra.fr/ft?id=%7B7858309A-82DB-4D52-9B5C-966887EBDEC3%7D
IPCC (2011) Renewable energy sources and climate change mitigation. special report of the
intergovernmental panel on climate change. In: Edenhofer O, Pichs-Madruga R, Sokona Y,
Seyboth K, Matschoss P, Kadner S, Zwickel T, Eickemeier P, Hansen G, Schloemer S, von
Stechow C (eds). Cambridge University Press, Cambridge and New York. http://www.ipcc.ch
/pdf/special-reports/srren/SRREN_Full_Report.pdf
Kim S, Dale BE (2011) Indirect land use change for biofuels: testing predictions
and improving analytical methodologies. Biomass Bioenergy 35(7):32353240.
doi:10.1016/j.biombioe.2011.04.039
Land Matrix (2013) Land matrix newsletterJune 2013. http://www.landmatrix.org/media/
filer_public/2013/06/10/lm_newsletter_june_2013.pdf
Millennium Ecosystem Assessment. (2005) Ecosystems and Human Well-being: Synthesis.
Washington, DC, Island Press
Mueller ND, Gerber JS, Johnston M, Ray DK, Ramankutty N, Foley JA (2012) Closing yield
gaps through nutrient and water management. Nature 490:254. doi:10.1038/nature11420
Murphy R, Woods J, Black M, McManus M (2011) Global developments in the competition for
land from biofuels. Food Policy 36 (Suppl 1):S52S61. doi: 10.1016/j.foodpol.2010.11.014
Oladosu G, Msangi S (2013) Biofuel-food market interactions: a review of modeling approaches
and findings. Agriculture 3:5371. doi:10.3390/agriculture3010053
PCFISU (2011) What price resilience? Towards sustainable and secure food systems. The
Princes Charities International Sustainability Unit, London. http://pcfisu.org/wp-content/
uploads/pdfs/TPC0632_Resilience_report_WEB11_07_SMALLER.pdf
Pimentel D, Marklein A, Toth MA, Karpoff MN, Paul GS, McCormack R, Kyriazis J, Krueger
T (2009) Food versus biofuels: environmental and economic costs. Human Ecol 37:112.
doi:10.1007/s10745-009-9215-8
Rosillo-Calle F (2012) Food versus fuel: toward a new paradigmthe need for a holistic
approach. ISRN Renew Energy 2012:Article ID 954180. doi:10.5402/2012/954180
Rulli MC, Saviori A, and DOdorico P (2013) Global land and water grabbing. Proc Nat Acad
Sci 110(3):892897. http://www.pnas.org/content/110/3/892.short
Searchinger T (2010) Biofuels and the need for additional carbon. Environ Res Lett 5:024007.
doi:10.1088/1748-9326/5/2/024007
Slade R, Saunders R, Gross R, Bauen A (2011) Energy from biomass: the size of the global
resource. Imperial College Centre for Energy Policy and Technology & UK Energy Research
Centre, London. http://www.ukerc.ac.uk/support/tiki-download_file.php?fileId=2095
Smeets EM, Faaij AP, Lewandowski IM, Turkenburg WC (2007) A bottom-up assessment and
review of global bio-energy potentials to 2050. Prog Energy Combust Sci 33(1):56106
UK (2013) State of nature. http://www.rspb.org.uk/ourwork/science/stateofnature/index.aspx
van Vuuren DP, van Vliet J, Stehfest E (2009) Future bio-energy potential under various natural
constraints. Energy Policy 37(11):42204230
WBGU (2008) Future Bioenergy and Sustainable Land Use. London and Sterling, VA, Earthscan
Chapter 3
Advanced Biofuels: Economic Uncertainties,
Policy Options, and Land Use Impacts
Wallace E. Tyner and Farzad Taheripour
Abstract Second generation or cellulosic biofuels have potential to become a reli-

able source of renewable fuel. In this chapter we describe five major uncertainties
that currently inhibit the development of these biofuels: (1) future oil prices, (2)
feedstock availability and cost, (3) conversion technology efficiency and cost, (4)
environmental impacts, and (5) government policy. In each of these areas, there are
significant issues that impede development and commercialization of the second
generation biofuels industry. However, all of these uncertainties could be managed
if society were willing to pay the higher cost of cellulosic biofuels.
3.1Introduction
During the past decade production of first generation biofuels, mainly corn and
sugarcane ethanol and biodiesel from oilseeds, have increased rapidly in response
to government policies and market incentives around the world and in particu-
lar in U.S., Brazil, and European Union. In 2000, the global production of bio-
fuels was limited to about 17.5billion liters of ethanol and 0.8million liters of
oilseed biodiesel. The global ethanol and biodiesel production in 2012 are about
86 and 21.5billion liters, respectively. Development of these biofuels has raised
concerns and debates on their environmental and economic consequences. Much
of the debate has focused on induced land use change emissions due to cropland
expansion for biofuel production, implications of using food crops for biofuel
W. E. Tyner(*) F. Taheripour
Department of Agricultural Economics, Purdue University, 403 West State St.,
West Lafayette, IN 47907-2056, USA
e-mail: wtyner@purdue.edu
F. Taheripour
e-mail: tfarzad@purdue.edu

36 W. E. Tyner and F. Taheripour
production, and welfare impacts of biofuel production and policy. With respect to
these issues, second generation biofuels, which can convert cellulosic materials to
liquid fuel, have gained favor among politicians. Some cellulosic biofuel pathways
have the potential to generate more environmental gains compare to the first gen-
eration. In addition, they may have fewer negative impacts on food production and
prices. However, there is very little commercial production of these biofuels today
because they are not economic, and policies have not been deemed adequate to
stimulate investment. This chapter discusses these issues following a brief review
of evolution in global biofuel production and policy during the past decade.
3.2Background
Production of biofuels was limited until a decade ago. In 2000, biofuel production
was limited to 10.5billion liters of Brazilian sugarcane ethanol, 6.2billion liters
of corn ethanol produced in U.S., and 0.8million liters of oilseed biodiesel mainly
produced in the European Union. During the past decade these regions have sig-
nificantly increased their biofuel production, and several other countries have
launched biofuel production. The global ethanol and biodiesel production will
reach 86 and 21.5billion liters in 2012. Figure3.1 shows distributions of these
biofuels among leading biofuel countries. The U.S., Brazil, and European Union
are the top three ethanol producers, and European Union, U.S., and Argentina are
the main biodiesel producers.
In general, the biofuel industries were launched under a mix of support poli-
cies including subsidies, mandate, and trade barriers across the world. However,
biofuel mandates became the dominant support policy over time. Biofuels have
Fig.3.1Global biofuel production in 2012

3 Advanced Biofuels: Economic Uncertainties 37
been a part of U.S. energy policy since 1978 (Tyner 2008a). Brazil also has a long
history of promoting biofuels (Tyner 2008b), and Europe also has promoted biofu-
els. Biofuels were developed for three main reasons: (1) reduction in greenhouse
gas emissions (GHG); (2) upgrading national security and reduction in oil depend-
ency; and (3) improvement in income for farmers and rural areas. U.S. develop-
ment of these biofuels has not been without controversy, as debates have raged
concerning the extent to which biofuels actually reduce GHG emissions and on the
extent to which biofuels have contributed to increased food pricesthe food-fuel
debate (Abbott et al. 2008, 2011; Tyner 2012). Most of these debates have per-
tained to first generation biofuels, and much less so to second generation.
Second generation biofuels, also called cellulosic biofuels, can be produced
from a wide range of cellulosic feedstocks such as corn stover, switchgrass, mis-
canthus, tree crops and residues, and municipal solid waste. While all of these
feedstocks are possible sources of second generation biofuels, there is little or no
commercial production of second generation biofuels today. As was the case with
first generation biofuels, second generation biofuels cannot grow without gov-
ernment support. The biofuels industry in the U.S. and elsewhere can be viewed
as a government created industry. In the U.S., corn ethanol production began in
the early 1980s after the passage of the Energy Tax Policy act of 1978, which
included a subsidy for corn ethanol (Tyner 2008a, b). The subsidy continued at
different levels through the end of 2011. Today the major driver of biofuels devel-
opment is the Renewable Fuel Standard (RFS) (U.S. Congress 2007), which man-
dates certain levels of biofuels of different types each year as shown in Fig.3.2.
Conventional biofuel (mainly corn ethanol) should reach 56.8billion liters in
2015 and remain at this level until 2022. U.S production capacity is now close to
the 56.8billion liters mandate for 2015. EPA has recently increased the biodiesel
Fig.3.2U.S. renewable fuel standard

mandate level to 5.2billion liters, and capacity exists for that level as well. The
government subsidy for corn ethanol ended at the end of 2011, and the biodiesel
subsidy also has been in limbo. For corn ethanol, the RFS has not been binding
(i.e. production has been more than RFS) so far, but it has for biodiesel. The rea-
son is that the wholesale price of corn ethanol minus ethanol tax credit was less
than the wholesale price of gasoline until 2011, and now the wholesale price of
ethanol is less than the wholesale price of gasoline. That is not the case for bio-
diesel, so the mandate continues to bind for this biofuel. As long as crude oil price
remains at about $85 per barrel or higher, corn ethanol can survive and produce
according to the RFS, even without government subsidies. However, second gen-
eration biofuels are more expensive to produce with current prices and produc-
tion costs. According to the RFS schedule the U.S. is supposed to produce about
1.9billion liters of cellulosic ethanol in 2012 and expand it to 60.6billion liters
in 2022. As mentioned earlier, the US currently produces very little cellulosic bio-
fuel, and its future expansion path is very uncertain.
So what are the prospects for the development and deployment of second gen-
eration biofuels? That is the subject of this chapter, and it depends critically on
five areas of uncertainty (Tyner 2010a). The uncertainties are: (1) future crude oil
prices, (2) feedstock availability and cost, (3) conversion technology efficiency
and cost, (4) environmental and GHG impacts, and (5) government policy. We will
review each of these uncertainty areas in turn and then provide some overall con-
clusions on the prospects for second-generation biofuels.
3.3Future Oil Price
The price of crude oil is a key variable which directly affects the prices of biofu-
els and their profitability. Other factors being constant, the higher the price of crude
oil the higher the prices for cellulosic biofuels. To be economic without subsidies
or mandates, given the existing production costs, second generation biofuels likely
will need crude oil to be $130$150 or higher. Figure3.3 provides the current U.S.
Department of Energy crude oil price forecast out to 2040 (U.S. Department of
Energy 2013). In the reference case, crude oil price grows from about $100/bbl. in
2012 to $163/bbl. in 2040. In the high oil price case, it grows to $235/bbl., and in
the low price case it becomes $75/bbl. All of these prices are in real terms; that is,
excluding the effect of general inflation. The forecast range between $75/bbl. and
$235/bbl. is very high, but even the reference case forecast would be problematic for
investors in second generation biofuels, because at least for the next decade the price
of crude oil will remain under $125/bbl. With crude oil prices below $130/bbl., pro-
duction of cellulosic biofuel likely will not be profitable without government support.
What is driving the wide price forecast range? The high case assumes more
rapid economic growth globally and more limited new oil and other liquid fuel
discoveries. The low case assumes slower economic growth and more abun-
dant crude oil and other liquid fuel supplies. Recent developments in oil supply
Fig.3.3Average annual Brent spot crude oil prices in three projection cases, 19802040
increases might push us closer to the lower end. The International Energy Agency
predicts that the U.S will become the worlds largest oil producer around 2020
and that North America will become a net petroleum exporter by around 2035
(International Energy Agency 2012a). These trends would eventually tend to push
crude oil prices lower unless there are corresponding reductions in production
elsewhere. Also, the natural gas revolution in the U.S. will create additional energy
supplies that will, to some extent compete with crude oil (International Energy
Agency 2012b). The bottom line is that crude oil price is highly uncertain, but
even the reference case is problematic for second generation biofuels. It is likely
that some form of government intervention will be necessary to stimulate invest-
ment in these biofuels.
3.4Feedstock Supply and Cost
The cost of feedstock is another key variable which affects profitability of pro-
ducing cellulosic biofuel. Current estimates of biomass feedstock costs are much
higher than earlier estimates (Congressional Research Service 2010; National
Research Council 2011; Thompson and Tyner 2013). For years, the standard figure
used by DOE was $33 per dry metric ton delivered to the plant. Today that cost is
estimated at $83$147 per dry metric ton, more than three times earlier estimates.
Since feedstock cost is the major component of variable cost, this difference has
a major impact on cellulosic biofuel cost. Table3.1 provides a range of feedstock
costs for various biofuel cellulosic feedstocks (National Research Council 2011).
Column 2 of this table shows the feedstock cost estimate, and column 3 is the esti-
mate of what the biofuel producer could afford to pay for the feedstock on a break-
even basis. Revenue was calculated from an assumed 2022 crude oil price of $111
Table3.1Willingness to pay and willingness to accept for alternative cellulosic feedstocks

Feedstock Willingness to Willingness to Price gap $/dry Price gap
accept $/dry pay $/dry metric ton $/liter
metric ton metric ton
Corn stover 101 28 74 0.25
Alfalfa 130 29 101 0.35
Switchgrass in the Midwest 147 29 118 0.40
Switchgrass in Appalachia 110 29 82 0.28
Miscanthus in the Midwest 127 29 98 0.34
Miscanthus in Appalachia 116 30 86 0.29
Wheat straw 83 30 53 0.18
Short rotation woody crops 98 26 72 0.25
Forest residues 86 26 60 0.20
Source National Research Council (2011)
per barrel. Conversion yield was assumed to be 292l of ethanol per dry metric
ton of feedstock. All estimated capital and operating costs except feedstock were
backed out leaving what the biofuel producer could afford to pay for feedstock.
The fourth column is the price gap or the difference between the cost to supply
the feedstock and what processors could afford to pay, expressed in dollars per
metric ton. The last column is that same price gap expressed in dollars per liter of
ethanol. These figures include no government subsidy. The bottom line is that with
these feedstock prices and technologies available today, cellulosic biofuels would
not be economic and will not be launched without government supports. Indeed,
the last column of Table3.1 shows required subsidy per liter of ethanol for alter-
native feedstocks when the crude oil price is $111/bbl. The required subsidies are
estimated to range between $0.18 and $0.40 per liter of ethanol.
While the cellulosic feedstock prices vary significantly, the quantity of avail-
able feedstock is not an issue (National Research Council 2011; U.S. Department
of Energy 2011). All the major estimates indicate that there would be plenty of
biomass available to meet and exceed the RFS requirement for cellulosic biofuels.
Another issue for feedstocks is variability in feedstock supply. This issue
applies to all agricultural feedstocks, but may be more problematic for corn stover.
It is likely that corn stover will be harvested after the corn harvest. Experts sug-
gest that there is about a three week window after corn harvest for removing the
stover and putting it in storage. If we have wet weather during that period such
that harvest is limited and/or the stover must be baled wet, then the stover yield
will be less than expected. Supply variability raises issues for contracting for bio-
mass supply (Alexander et al. 2010). Contracts must work both for the farmer and
for the processor. Contracts will need to be designed to share risk between farm-
ers and processors such that both parties have incentives to enter into long term
contracts and abide by the contract terms. Contracts may be based on acreage, ton-
nage, quality, and other factors. The bottom line is that while availability of bio-
mass is not an issue for biofuels production, feedstock cost and feedstock supply
variability could be significant impediments.
3.5Conversion Technology
There are two major conversion pathways for biofuel production: biochemical
conversion and thermochemical conversion. In addition, there are hybrid processes
that are partly biochemical and partly thermochemical. The biochemical path-
way normally results in ethanol as the biofuel product. The thermochemical path-
way normally produces green diesel, bio-gasoline, or jet fuel. These products are
sometimes called drop-in fuels as they are closer in composition to the existing
fossil based fuels and unlike ethanol can be more easily integrated into the exist-
ing fuel supply chain. Currently, the U.S. ethanol industry is faced with a restric-
tion, known as the blend wall. Since the early days of the U.S. ethanol industry,
the regulated maximum ethanol content for standard vehicles has been 10%.
Standard vehicles were designed to handle blends only up to that level. Currently,
U.S. annual gasoline consumption is about 500billion liters. This means the total
consumption of ethanol, with current mix of vehicles, will be limited to 50bil-
lion liters, which is about the current U.S. ethanol production. Consuming ethanol
beyond this level will be constrained by the blend wall. Unlike ethanol, drop in
fuels do not have blending limits. They can be blended in much higher percentages
than is the case for most ethanol use in the U.S. Also, they can be shipped in pipe-
lines, which is not possible for ethanol.
Earlier in the biofuels era, most of the research and development inter-
est was focused on the biochemical pathway. However, in recent years, much
more interest has emerged on the thermochemical pathway or hybrid pathways
(Tyner 2010b). Drop in fuels are seen as more attractive for many of the reasons
described above.
For the conversion technologies, it is known that biofuels can be produced via
either the biochemical or thermochemical pathways. The question is at what cost?
Since there are no commercial plants, any cost estimate is uncertain. Research
continues on both biochemical and thermochemical process as well as several
hybrid processes. Major breakthroughs are possible at any point in the future.
The estimates for conversion costs vary among alternative conversion pathways
and feed stock. The capital cost for the thermochemical pathway is higher than
the biochemical, $0.30 per liter versus $0.15 per liter (Taheripour et al. 2011). On
the other hand, the variable costs (excluding feedstock) are higher for biochemical
pathways. For example, the variable cost of converting corn stover to biofuel using
a biochemical process is estimated to be about $0.37 per liter. The correspond-
ing cost for the thermochemical is about $0.13 per liter (Taheripour et al. 2011).
If feedstock is $110 per metric, and ethanol yield is 292l per metric ton, then
feedstock cost would be $0.38 per liter. This means that the total conversion cost
(including capital, operating, and feedstock costs) for converting cellulosic materi-
als to biofuel is around $0.88 per liter plus or minus a few cents variation among
alternative pathways and feedstock. This is significantly higher than the total con-
version cost of corn ethanol which is about $0.65 per liter at current prices (Iowa
State University 2012).
3.6Environmental Issues
Biofuels were developed to mitigate GHG emissions. For example, prior to 2007,
the general consensus was that corn ethanol can reduce GHG emissions a bit more
than 20% after considering all emissions generated through the production process
of ethanol and its consumption (Wang et al. 1999; Farrell et al. 2006). However,
early analyses ignored the consequences of biofuel induced land changes for GHG
emissions. Once early estimates of GHG emissions due to land use change were
included, it appeared that corn ethanol was more GHG intensive than gasoline
(Searchinger et al. 2008). These authors estimated that producing corn ethanol gen-
erates more than 100g CO2 e/MJ emissions due to induced cropland expansion
for required corn production. However, the more recent studies find that the early
estimates have overstated the induced land use emissions due to biofuels (Hertel
etal. 2010; Tyner and Taheripour 2013). For example, Taheripour and Tyner (2013)
estimated that producing corn ethanol generates about 1323g CO2 e/MJ emis-
sions due to cropland expansion for required corn production, depending on the
implemented land sue emission factors and assumptions on land use carbon fluxes.
While the more recent estimates for induced land use change for the first generation
biofuels are usually lower than their earlier estimates (Wicke et al. 2012), they are
significantly different from zero. This indicates that first generation biofuels do not
contribute significantly to reducing GHG emissions, if we take into account their
related induced land use change emissions. As technologies improve, it is possible
that the GHG emission profiles for first generation biofuels will improve as well.
The first generation biofuels could cause other environmental concerns as well.
Corn production is intensive in fertilizer and chemicals, and increased corn acre-
age likely would lead to higher fertilizer and chemical runoff and soil erosion
(National Research Council 2011). These effects would lead to a reduction in
downstream water quality. There also has been some concern regarding local air
quality issues associated with ethanol production.
While many studies evaluated environmental impacts of the first generation
biofuels, only a few attempts have been made to assess these impacts for sec-
ond generation biofuels. In general, second generation biofuels are believed to
have more positive environmental impacts. Perennial grasses like miscanthus and
switchgrass need fewer nutrients added and do a better job of preventing soil ero-
sion. In addition, producing dedicated crops on marginal lands can increase their
carbon sequestration capabilities (Anderson-Teixeira et al. 2009). As mentioned
earlier, cellulosic biofuels also can be produced from crop or forest residues. If
biofuels are produced from these materials, then land use implications will be
around zero as shown in Taheripour and Tyner (2013).
However, if cellulosic biofuels are produced from dedicated crops, their land
use change emissions will not be zero. Taheripour and Tyner (2013) have esti-
mated induced land use emissions for several biofuel pathways under alternative
assumptions of soil carbon emissions factors and assumptions on land use change
carbon fluxes. The core of their results is shown in Table3.2. This table shows
induced land use change emissions for seven biofuel pathways, corn ethanol
Table3.2Estimated induced land use emissions for alternative biofuels (g CO2 e/MJ)
Feedstock Biofuel Without CP-EF With CP-EF
WH CARB TEM WH CARB TEM
Corn Ethanol 12.9 15.1 17.0 15.5 18 22.6
Miscanthus Bio-gasoline 6.1 7.1 7.3 18.1 19.4 25.6
Switchgrass Bio-gasoline 21.4 24.9 23.4 43.7 47.6 57.0
Miscanthus Ethanol 5.8 10.1 10.1 15.7 25.4 32.3
Switchgrass Ethanol 20.3 35.5 33.1 38.2 63 74.0
Source Taheripour and Tyner (2013)
plus 6 cellulosic biofuel pathways. The induced land use changes due to produc-
ing required feedstock for each pathway are converted to GHG emissions using
three different published land use change emission factors (EFs). The imple-
mented EFs are: Woods Hole Emission Factors (WH), emission factors developed
by California Air Resources Board experts (CARB), and a set of emission factors
developed using a Terrestrial Ecosystem Model (TEM). Finally, for each emis-
sion factor calculations are done under two cases. The first case assigns zero EFs
to changes in converted marginal land (cropland pasture) to dedicated crops. In
Table3.1 this case is shown under the title of Without CP-EF. The second case
assumes emission EF for cropland pasture in each agro-ecological zone is equal
to the half of the emissions factor associated with the pasture land in than zone. In
Table3.1 this case is shown under the title of With CP-EF.
The Table3.1 shows that unlike the common belief, producing biofuels from
dedicated crops generates induced land use change emissions. The results also
indicate that the land use change emissions vary significantly among alternative
sets of EFs and biofuel pathways. When we assume converting cropland pasture
converted to dedicated crops does not generate carbon emissions, then produc-
ing ethanol and or bio-gasoline from miscanthus generate the lowest induced land
use emissions. However, in the counterpart case, when non-zero emission factors
are assigned to converted cropland pasture, then corn ethanol generates the low-
est emissions. This means that second generation biofuels produced from dedi-
cated crops could actually generate more induced land use change emissions than
first generation. Finally, this table indicates that switchgrass generates the higher
rate of land use emissions (even larger than corn), regardless of the type of imple-
mented EF in both cases of with and without CP-EF.
These results show another set of uncertainties associated with the second gen-
eration biofuels. These results show that at the end of the day, the limiting resource
is land, and second generation feedstocks (except residues) do require land. For
residues such as corn stover, there is an unresolved question of the extent to which
residue removal reduces soil carbon stock, and thus has adverse GHG impacts. For
land using perennial crops like switchgrass, the land required often would come
from the livestock sector. If more land is used for biofuel feedstocks, less would
be available for cattle grazing, hay production, etc. Thus, it is not entirely true that
second generation biofuels do not generate land use change emissions. In addition,
the second generation biofuels, if produced from dedicated crops, will compete
with food crops in the market for land. Hence, the issue of food versus fuel will
not disappear, if the second generation biofuels are produced from dedicated
crops. Most of the corn produced is actually used for animal feed, not for direct
human consumption. So corn use for ethanol competes with food use primarily via
the livestock sector and meat consumption and prices. The same is true, perhaps to
a lesser degree, for second-generation dedicated energy crops.
3.7Government Policy
The U.S. corn ethanol industry, Brazilian sugarcane ethanol industry, and
European Union biofuels industry (mainly biodiesel) were all industries created
via the support from governments. Support for these industries generally began in
the 1980s. In all three regions, support initially came in the form of government
subsidies. However, over time governments have evolved towards implementing
mandates in lieu of subsidies (Tyner 2008a, b). As biofuels production increased,
the government cost of the subsidies became a significant burden. The cost of
mandates is off-line; that is, the cost is imposed on consumers via the mandate,
and does not show up in government budgets.
In the early years, there was strong political support for government support
of biofuels. Agricultural groups had a disproportionate amount of political power,
and the biofuel support was not questioned. In recent years, however, the politi-
cal support in the U.S. has waned. Sectors affected by higher food pricesfood
manufacturers and retailers, restaurants, etc. have come out against biofuels sup-
port. Even the agricultural livestock producers have broken rank with the rest of
the agricultural sector in opposing biofuels. So the reality today is that biofuel sup-
port is less certain than it was in the early years.
The major biofuel support policy today in the U.S., Brazil, and the European
Union is some sort of mandate or target for biofuels use. In the U.S., the mandate
is the RFS. As shown in Fig.3.1, it requires 136.3billion liters ethanol equivalent
of biofuels to be blended by 2022. The RFS that can be filled by corn ethanol in
2015 is 56.8billion liters, and U.S. production capacity is about to that level today.
For cellulosic biofuels, the RFS is 60.6billion liters ethanol equivalent. There is
another category called other advanced biofuels (with 15.1billion liters target for
2022) that can be met by sugarcane ethanol among other possibilities. The final
mandate is for 3.8billion liters of biodiesel.
To enforce the mandates the Environmental Protection Agency (EPA) uses a
procedure which tracks biofuel and consumption. In this process, the EPA issues
a unique Renewable Identification Number (RIN) when a batch of biofuel is pro-
duced or imported. Each obligated party (blender) has a quota for blending each
year that is based upon their projected fuel market share. At the end of the year,
obligated parties (blenders) must present RINs equivalent to their quota to the EPA
to prove that they meet their annual biofuel obligations. At the end of each year, if
a blender does not have enough RINs to meet its obligation, it will be punished a
fine. RINs can be traded in market place. Hence, if a blender needs RINs to meet
its obligation, it can buy from the market and if it has extra RINs it can save it for
future use or sell it to the market.
While the RFS has defined annual targets for each biofuel, the law allows the
EPA to waive them for cellulosic biofuels, under some certain circumstances. EPA
has waived the cellulosic biofuel mandates every year so far, mainly because no
cellulosic biofuel is produced commercially. The cellulosic portion of the RFS has
a special provision in which for any year in which any part of the cellulosic RFS is
waived, it is possible to buy a credit from EPA instead of actually blending cellu-
losic biofuel. A company can buy a credit from EPA and buy an advanced biofuel
blending RIN from another company and use the combination of the EPA credit
plus the advanced biofuel (e.g. sugarcane) RIN to satisfy the cellulosic blending
mandate (U.S. Congress 2007). The cost of the credit from EPA is determined by
the difference between the previous year wholesale gasoline price and a base gaso-
line price (about $0.80 per liter) adjusted for inflation since 2007. In other words if
the 2012 value is $0.85, and the 2012 average wholesale gasoline price was $0.70,
then an obligated blender could purchase from EPA a credit for $0.15 per liter. The
RIN would have to be purchased on the open market, and might have a value of
around $0.10 per liter. Thus, an obligated blender could blend cellulosic biofuel
or purchase the credit plus RIN for $0.25 per liter and meet the obligation in that
way. If the cost of the cellulosic biofuel is about $1 per liter, and the wholesale
price of gasoline is $0.70, the difference is $0.30 per liter. The obligated blender
would have the choice of blending at a cost of $0.30 or purchasing the credit and
RIN for $0.25. Clearly, under these assumptions, it would be more attractive to
purchase the credit and RIN instead of purchasing cellulosic biofuel. This means
that the mandate is not really a mandate in reality. This out-clause creates huge
uncertainty for potential investors in cellulosic biofuels. Cellulosic biofuels are not
competitive on the open market. The hope and expectation was that the govern-
ment mandate would create the market, but with this out-clause, it clearly does
not, at least under the assumptions used here. Thus, it will be difficult to attract
private sector investment into the industry.
The above analysis applies anytime EPA waives any part of the cellulosic RFS.
In reality, EPA has waived most of the cellulosic mandate in every year it has
existed. It will be necessary for EPA to waive part of the mandate all the way out
to 2022 because the industry, even if it became attractive, could not possibly grow
as fast as the mandate.
There have also been requests to waive the corn ethanol mandate with the most
recent being due to the 2012 U.S. drought (Tyner et al. 2012). The criterion EPA
is required to use in this case is economic harm caused by the RFS. In November
2012 EPA ruled that the 2013 corn ethanol mandate would not be waived.
However, the possibility of a waiver at any time also creates uncertainty for the
industry.
Another major impediment to growth of the ethanol industry in the U.S. is what
is called the blend wall. Most gasoline in the U.S. is blended at 10% with etha-
nol. The U.S. currently consumes about 500billion liters of gasoline type fuel.
Fig.3.4Ethanol market in
the presence of blend wall
Blending at 10% means that the maximum amount of ethanol that can be blended
is about 50billion liters. A very small amount of E85 (85% ethanol) is sold for
use only in flex fuel vehicles, but these vehicles represent a small portion of the
U.S. vehicle fleet. While EPA has approved moving to 15% blends, very little
progress has been made in implementing the decision. The EPA approval was for
vehicles built since 2001, and that amounts to about 2/3 of the vehicle fleet (Tyner
et al. 2010). If a service station were to change to E15 (15% ethanol), they would
lose 1/3 of their customers. E15 also cannot be used in any small engines such as
marine engines, lawn mowers, chain saws, etc. This means that the new rule has
not been able to remove the exiting blend wall so far.
The economics of the blend wall are illustrated in Fig.3.4. The kinked bold
line represents the demand curve in the presence of blend wall. When we reach
the aggregate blending limit, there is no more room for additional ethanol in the
market place. We are then in a position of too much ethanol (production capacity)
chasing too little market (the blend wall). At that point, the ethanol price becomes
its breakeven price with the corn price. In fact, we see that relationship existing
today, and the recent historical link between gasoline and ethanol is largely bro-
ken. Because of the blend wall, ethanol is largely priced today on corn.
The blend wall has very important implications for cellulosic ethanol. Even if
the blending limit of 15% can be applied more broadly, that still limits the total
ethanol blending to about 72billion liters. There is no place to put a significant
amount of cellulosic ethanol. That is one reason for the increased attention to ther-
mochemical conversion and drop-in biofuels.
Finally, there is one government policy option that merits serious attention and
reduces uncertainties in markets for cellulosic biofuels significantly. It is called a
reverse auction. The U.S. Air Force and Navy have expressed a strong interest in
biofuels (U.S. Air Force 2010). The military could use a reverse auction to procure
biofuels. In so doing, it would specify the fuel properties, delivery location, and
quantity to be delivered each year for about 15years. A long term contract would
be necessary to gain private sector participation in the bidding. Once the call for
bids was out, private companies could secure provisional feedstock contracts with
farmers and submit a bit. The lowest bid from a qualified bidder wins the contract.
This approach reduces the adverse impact of the uncertainties described above in
several ways. First, it sets the fuel price, so it does not really matter what happens
to oil price in the future. Second, feedstock uncertainties are handled because bid-
ders would have provisional contracts with farmers that would be executed if they
won the competition. Third, presumably companies would not be bidding unless
they were confident of the technical and economic dimensions of their conver-
sion technology. Fourth, government policy and support would not matter because
this becomes a private contract between the military and the company. Thus, the
reverse auction could be a mechanism to get the cellulosic biofuel industry mov-
ing. In a sense, the difference between the winning bid price and the projected fos-
sil fuel price for the same commodity becomes the implicit subsidy. This implicit
subsidy is a least cost option since it is market determined. At this point, the mili-
tary does not have Congressional approval to use this policy mechanism, so any
movement in this direction is blocked.
3.8Conclusions
Second-generation (cellulosic) biofuels have potential to become a reliable source

of renewable fuel. In this chapter we have described five major uncertainties that
currently inhibit the development of these biofuels: (1) future oil prices, (2) feed-
stock availability and cost, (3) conversion technology efficiency and cost, (4) envi-
ronmental impacts, and (5) government policy. In each of these areas, there are
significant issues that impede development and commercialization of the second
generation biofuels industry. However, all of these uncertainties could be managed
if society were willing to pay the higher cost of cellulosic biofuels.
References
Abbott P, Hurt C, Tyner WE (2008) Whats driving food prices? Farm foundation report. https:
//www.farmfoundation.org/webcontent/Farm-Foundation-Issue-Report-Whats-Driving-Food-
Prices-404.aspx. Accessed 6 May 2013
Abbott P, Hurt C, Tyner WE (2011) Whats driving food prices in 2011? Farm foundation report.
http://www.farmfoundation.org/webcontent/Whats-Driving-Food-Prices-in-2011-1742.aspx.
Accessed 6 May 2013
Alexander C, Ivanic R, Rosch S, Tyner WE, Wu S, Yoder J (2010) Contract theory and implica-
tions for perennial energy crop contracting. Energy Econ 34:970979
Anderson-Teixeira K, Davis S, Mater M, Delucia E (2009) Changes in soil organic carbon under
biofuel crops. GCB Bioenerg 1:7596
Congressional Research Service (2010) Cellulosic ethanol: feedstocks, conversion technologies,
economics, and policy options. Washington, D.C., Contract No: R41460. http://www.national
aglawcenter.org/assets/crs/R41460.pdf. Accessed 6 May 2013
Farrell AE, Plevin RJ, Turner BT, Jones AD, OHare M, Kammen DM (2006) Ethanol can con-
tribute to energy and environmental goals. Science 311:506508
Hertel T, Golub A, Jones A, OHare M, Plevin R, Kammen D (2010) Effects of US maize etha-
nol on global land use and greenhouse gas emissions: estimating market-mediated responses.
Bioscience 60:223231
International Energy Agency (2012a) World energy outlook executive summary. Paris, France.
http://www.iea.org/publications/freepublications/publication/English.pdf. Accessed 6 May 2013
International Energy Agency (2012b) Golden rules for a golden age of gas. Paris, France. http://ww
w.worldenergyoutlook.org/media/weowebsite/2012/goldenrules/weo2012_goldenrulesreport.pdf
Iowa State University (2012) Ethanol profitability 2012. Available from: http://www.agmrc.org/
renewable_energy/
National Research Council (2011) Renewable fuel standard: potential economic effects of U.S.
biofuel policy. National Academy of Sciences. http://www.nap.edu/openbook.php?record_
id=13105&page=1. Accessed 6 May 2013
Searchinger T, Heimlich R, Houghton RA, Dong F, Elobeid A, Fabiosa J, Tokgoz S, Hayes D, Yu
T (2008) Use of U.S. croplands for biofuels increases greenhouse gases through emissions
from land use change. Science 319:12381240
Taheripour F, Tyner WE (2013) Induced land use emissions due to first and second generation
biofuels and uncertainty in land use emission factors. Econ Res Int, vol 2013, http://www.hin
dawi.com/journals/econ/2013/315787/
Taheripour F, Tyner WE, Wang MQ (2011) Global land use changes due to the U.S. cellulosic
biofuel program simulated with the GTAP model. Argonne National Laboratory and Purdue
University. http://greet.es.anl.gov/files/luc_ethanol. Accessed 6 May 2013
Thompson J, Tyner WE (2013) Corn stover for bioenergy production: Cost estimates and farmer
supply response. Biomass Bioenergy (in press)
Tyner WE (2008a) The US ethanol and biofuels boom: Its origins, current status, and future pros-
pects. Bioscience 58:646653
Tyner WE (2008b) The global impacts of US and EU biofuels policies. In: Auurbier P, van de
Vooren J (eds) Sugarcane ethanol: contributions to climate change mitigation and the envi-
ronment. Wageningen Academic Publishers, Wageningen, pp 181197
Tyner WE (2010a) Why the push for drop-in biofuels? Biofuels 1:813814
Tyner WE (2010b) Cellulosic biofuels market uncertainties and government policy. Biofuels
1:389391
Tyner WE (2012) Biofuels and agriculture: a past perspective and uncertain future. Int J Sustain
Devel World Ecol 19:389394
Tyner WE, Dooley F, Viteri D (2010) Alternative pathways for fulfilling the RFS mandate. Amer
J Agric Econ 92:18
Tyner WE, Taheripour F, Hurt C (2012) Potential impacts of a partial waiver of the ethanol blend-
ing rules. Farm foundation report. http://www.farmfoundation.org/news/articlefiles/1841-Pur-
due%20paper%20final.pdf. Accessed 6 May 2013
Wang M, Saricks C, Santini D (1999) Effects of fuel ethanol use on fuel-cycle energy and green-
house gas emissions. Center for Transportation Research, Energy Systems Division, Argonne
National Laboratory. http://www.transportation.anl.gov/pdfs/TA/58.pdf. Accessed 6 May 2013
Wicke B, Verweij P, van Meijl H, van Vuuren DP, Faaij APC (2012) Indirect land use change:
review of existing models and strategies for mitigation. Biofuels 3:87100
U.S. Air Force (2010) Air force energy plan 2010. http://www.safie.hq.af.mil/shared/media/
document/AFD-091208-027.pdf. Accessed 6 May 2013
U.S. Congress (2007) Energy Independence and Security Act of 2007, in H.R. 6, 110 congress,
1st session
U.S. Department of Energy (2011) U.S. billion ton updatebiomass supply for a bioenergy
and bioproducts industry. http://www.usbiomassboard.gov/pdfs/eaton_tac_q3_2011.pdf.
Accessed 6 May 2013
U.S. Department of Energy (2013) Annual energy outlook 2013: early release. Washington, D.C.
2012. http://www.eia.gov/forecasts/aeo/er/index.cfm. Accessed 6 May 2013
Chapter 4
Algae Farming and Its Bio-Products
Gal Hochman and David Zilberman
Abstract Many expect algae to contribute to food, feed, health, and fuel, as well
as to remove or transform pollutants in water or air. But what did we really achieve
after several decades of research and development? What ended up being commer-
cialized and consumed in large volumes? We try and shed light on these questions
by surveying and assessing the current state of algae uses.
4.1Introduction
Increased concern about climate change, energy security and non-renewability

associated with the dependence on fossil fuel has led to investigate alternative
sources of renewable fuels. Algae have been seen as an attractive feedstock for
biodiesel because of their capacity to produce oil under stresses, frequently using
waste products as nutrients (Sheehan et al. 1998). Wiley and Campbell (2011) and
Li et al. (2008) presented some of the processes that are considered for utiliza-
tion of algae as a fast growing feedstock for biofuel. But Gallagher (2011) argues
that productivity, capital costs, technical constraints, and uncertainty, constrain the
economic viability of algae as biofuel. While future research will identify new cost
reducing strategies to utilize algae for fuel production, algae has a large economic
potential in other applications some of which can complement the use of algae as
G. Hochman(*)
Department of Agriculture, Food, and Resource Economics, School of Environmental and
Biological Sciences, Rutgers University, 55 Dudley Road, Cook College, New Brunswick,
NJ 08901, USA
e-mail: gal.hochman@rutgers.edu
D. Zilberman
Department of Agricultural and Resource Economics, University of California,
207 Giannini Hall, Berkeley, CA 94720, USA
e-mail: zilber11@berkeley.edu

50 G. Hochman and D. Zilberman
feedstock for fuel. This chapter overviews a wide range of applications of algae
that has or likely to have significant economic potential. The development of algae
as feedstock for biofuel will benefits from technological breakthroughs in other
applications, and understanding these other application is important in developing
sustainable and diversified algae products and technologies.
Algae are large and diverse group of organisms, typically autotrophic organ-
isms that produce complex compounds such as lipids, carbohydrates, and proteins,
using simple substances located in their surroundings. Although most algae are
photosynthetic plant like that lack the distinct cell and organ types present in
land plants, some produce energy from uptake of organic carbon.
Algae have raised much hope and scholars have argued that there is much
potential of using algae for wastewater treatment, as well as feed food health and
fuel production. Many barriers, however, exist for these processes to become eco-
nomically viable and environmentally friendly. The development of molecular
genetics (biotechnology) raised much hope in improving algae strains and making
them commercially viable but also brought much fear, because large portions of
the population are suspicious and oppose this technology. Biotechnology is a field
of applied science and technology, which employs living organisms and their sub-
cellular components for industrial applications and environmental management.
Biotechnology makes use of viruses, bacteria, yeasts, fungi, algae, plant and ani-
mal cells, and enzymes as components of industrial processes.
This chapter survey current developments of algal bio-products. The chapter
surveys the economic performance of the algae industry thus far and the economic
potential of alternative product lines. The analysis, especially the economic analy-
sis and the cost estimates, are based on interviews with scientists and business-
men. But because of confidentiality, the findings are not attributed directly to their
sources. The list of sources, however, is provided in Appendix A.
4.2Algae Farming
Existing patterns of use of algae are insightful and useful for future projections. Still,
present success stories are only the tip of the iceberg. Much more research and knowl-
edge are needed for full societal gains from the economic potential of algal products.
When surveying algae, one has to distinguish between two major signsmicroalgae
and microalgae. While macroalgae are basic seaweeds, microalgae are microscopic
organisms, which make up the worlds phytoplankton and form rapidly growing pop-
ulations in water when supplied with the necessary nutrients for their culture.
4.2.1Macroalgae
Macro-algal grow on rocky substrates, forming multilayer perennial vegetation cap-

turing almost all available photons. The macro-algal maximum productivity is 10
times higher that of plankton population (Carlsson and Bowles 2007). The maximum
4 Algae Farming and Its Bio-Products 51
chlorophyll content corresponds to an algal biomass of about 10kg per m2 (Luning

and Pang 2003).
About 200 species of macro-algae are used around the globe, where 10 of
which are intensively cultivated (Luning and Pang 2003),
Brown algae: Laminaria japonica and Undaria pinnatifida;
Red algae: Porphyra spp., Eucheuma spp., Kappaphycus spp., and Gracilaria
spp., and;
Green algae: Monostroma spp., and Enteromorpha spp.
Suitable macroalgal species used for large-scale cultivation include species of
Alaria, Corallina, Cytoseria, Ecklonia, Egregia, Eucheumia, Gracillaria, Laminaria,
Macrocystis, Pterygophora, and Sargassum (Carlsson and Bowles 2007).
The world production of macroalgae for commerce amounts to $5.56billion
a year (McHugh 2003; Pulz and Gross 2004). Many types of dried seaweed are
being used as food products, mainly in salads and for seasonings, with currently
species of Rhodophhyta and Phaeophyta used industrially to produce 7.58mil-
lion tons of wet seaweed annually. While the food industry is estimated to generate
$5billion a year, a further $600million is estimated to have been generated from
hydrocolloids extracted from the cell wall of the macro-algae at an average value
of about $10,900 a ton. Sales of one of the dried kelp, called Nori, are estimated to
be $1billiona high value product worth $16,000 a ton. While Nori is consumed
mostly in Japan and the Far East, its consumption is spreading to the West and it is
now being produced and consumed in California. Much of the Nori in the market
is harvested from the sea, but there are substantial efforts in cultivating it through
mariculture.
But the economic value of the macroalgae derivatives goes much beyond their
usefulness as food products. Agar is the most diversely used macroalgae deriva-
tive with substantial worldwide sales. Agar is a class of vegetable gums that is
derived from the two varieties of seaweedGelidium and Gracilaria. It is a very
strong gelling agent with unique properties. Agar is not poisonous to humans, has
no nutritional content, does not rot, and can absorb liquids and swell. The gel it
generates may survive a wide range of temperatures; is indigestible by most bac-
teria; and is very elastic, resilient, and clear (Chapman 1970; Renn 1984). Agar
and its derivatives have a wide variety of uses. The value of agar products varies
substantially according to their quality.
The relatively low-value agar is used in the food industry (emulsifier, gelling
agent, and preserving agent) as a component of many laxatives and as an impres-
sion material (especially in dentistry). The medium-value use of agar is in bacte-
riological and microbiological applications. The highest-value derivative of agar
is called agarose and is used in a microbiological genetic-engineering application.
The demand for agar and agarose is continuing to grow, and there is much
interest in alternative sources of agar and some experimentation with domestica-
tion of it. Agar served as an example to the larger market for algae derivatives and
versatility of products of different quality that can be produced from individual
algae. It also serves to emphasize that this is a time where domestication of sea-
weed production attracts much interest.
Carrageenan and alginates are two other macroalgae derivatives much used
as gums, emulsifiers, and gels. Annual sales of these products are in the hun-
dreds of millions of dollars. According to McHough (2003), the market value of
carrageenan is $240 million, that of alginate is $213 million, and that of agar is
$137million. Producing methane from macro-algae has also been discussed in the
literature; however, it has been argued that algal biomass production, as a stand-
alone product, is not economically viable (Chynoweth 2002).
4.2.2Microalgae
There are more than 8,000 microalgae species which are divided into four types:
cynaobacteria (blue-green algae), rhodophytes (red algae), chlorophytes (green
algae), and chromophytes (all other algae). Each of these types contains hundreds of
species. Each species may be thousands of genetically distinct strains. Only a small
fraction of these varieties have been studied for possible beneficial use, and there
is much ignorance and uncertainty regarding the behavior and properties of most
micro-algae species. The most frequent used micro-algae include Cyanophyceae
(blue-green algae), Chlorophyceae (green algae), Bacillariophyceae (including dia-
toms), and Chrysophyceae (including golden algae).
The world market value of micro-algae has been estimated at $56.5billion,
out of which about 2.5billion dollars have been generated by the health food sec-
tor, 1.5billion dollars from the production of docosahexanoic acid (DHA) and
700million dollars from aquaculture (Pulz and Gross 2004).
4.3Commercial Uses of Algae
Much interest is currently expressed in production of algal biomass. Present calcu-

lations of production costs of algal biomass suggest that with current technology
it not economically viable as a stand-alone product, although several studies have
argued that co-production of algal biomass may become viable under certain sce-
narios (Carlsson and Bowles 2007; Lundquist et al. 2010; Hochman et al. 2013).
However, co-producing algal biomass limits the scale of energy production to the
profitable application (Reith 2004). Lundquist et al. (2010) argued that co-produc-
ing algal biomass with wastewater treatment is less limiting, but that co-producing
algal biomass with value added products such as astaxanthin and -carotene does
significantly limit algal biomass production.
Algae can be used to produce raw material for co-firing to produce electricity,
liquid fuel production via pyrolysis and thermochemical liquefaction (bio-oil), or
biomethane generation through fermentation. While these processes cannot, yet,
compete with fossil fuels and their heating value is low with 29MJ/kg compared
with 42MJ/kg (Miao et al. 2004), algae energy content and heating value is higher
than that of other biomass feedstock. Major limitations of commercialization of

algae biofuels include (1) algae mass culture oil content; (2) harvesting methods;
(3) separation techniques; and (4) supply of CO2 and other nutrients (Miao et al.
2004; Lundquist et al. 2010).
However, small-scale cultivation and industrial scale production of microalgae
has evolved in the last few decades. Several substantial applications have been
established. These applications include:
4.3.1Waste Water Treatment
Micro- and macro-algae can be used to sequester, remove, or transform pollut-

ants including excess heavy metal, nutrients, and xenobiotics from wastewater, or
CO2 from exhausts. The yield derived from this process can be algal biomass used
to produce chemicals, biofuels, bio-oil, and biogas as co-products (Munoz and
Guieysse 2006).
Many studies and several commercial facilities have demonstrated the viability
of microalgae in sewage treatment (for example, Oswald 1987a, b). Oxygen produc-
tion by microalgae for waste oxidation by bacteria in ponds is generally recognized.
There are also promising results demonstrating algae contribution in enhancing sedi-
mentation, disinfection, nutrients, and in removing heavy metal and organic toxins.
Oswald estimates the savings associated with the use of algae (in place of electricity)
for oxygen production in sewage ponds to be between $3,300 and $14,000 (1985)
per hectare (based on an energy price of 10cents per kilowatt-hour).
The algae biomass produced in sewage pools can be used to produce energy
by fermentation. There have been several large-scale experimentations in the com-
bined use of algae for waste management and energy production, and a combined
system seems especially appropriate and economical to locations with expensive
and scarce energy resource.
The use of alga for sewage oxidation (including use of the resulting biomass for
energy production or other economic purposes) is likely to increase substantially
as energy prices increase and more knowledge about the technology becomes
available through experience. Increased productivity is an important factor in
determining the future of the technology and will influence its fate.
There is substantial demand for technologies capable of removing chemicals
from bodies of water. For example, there has been extensive search in California
for technologies capable of ridding water of selenium and other minerals and tox-
ins. The volume of the problem is immense, and hundred of millions of dollars are
allotted annually to waste treatment. Similar problems occur elsewhere and sug-
gest a good area for future applications for algal use (given that through research
algal technologies can provide effective and economical solutions).
There are many links and dependencies between microalgae for waste research.
There is much potential for economic gain combining the use of algae for
waste management and other activities (Shelef 1982). Many of the technologies
developed in the use of microalgae for waste management systems are appropri-
ate for algal utilization in other production activities. Insights regarding the use of
microalgae for waste management have commercial value and can be sources of
income. There are markets for expertise in water and waste management.
4.3.2Fine Chemicals
Like agar, algal products vary substantially in price and value, according to their
use and refinement.
-Carotene is a metabolite with a wide range of commercial applications. It is
used as a food coloring (with a major application in providing the yellow color to
margarine), as a good additive to enhance the color of the flesh of fish and the yolk
of eggs, and to improve the health and fertility of grain-fed cattle (see survey by
Borowitzka and Borowitzka 1987).
Until the early 1980s, commercial production of -carotene was synthetic, and
Hoffman Laroche had a virtual monopoly on the production and marketing of
-carotene. During the 1970s, researchers (Borowitzka and Brown 1974;
BenAmotz and Avron 1980) realized that, under nutrient stressed, high salt, and
highlight conditions, the microalgae, Dunaliella salina, accumulates up to 14%
of dry weight as -carotene. This discovery led to commercial derivation of natural
-carotene from this organism.
In the 1980s, the price of extracted and purified natural -carotene was much
higher than that of synthetic -carotene ($1,000 to $2,000 per kg for natural versus
$400 to $800 per kg for synthetic), reflecting the preference consumers and buyers
have for natural products. Even though the price difference between natural and syn-
thetic declined in the future as the supply of natural -carotene increased, the differ-
ences continued and the natural product always fetched the higher price. Moreover,
natural -carotene has physical properties that make it superior to synthetic. In particu-
lar, natural -carotene is fat-soluble. -carotene, as well as some other carotenoids, are
touted to be anti-carcinogenic and are effective in controlling cholesterol and in reduc-
ing risks of heart disease (Nishino et al. 2002). If substantiated, these desirable medi-
cal properties increase even more the demand and desirability of natural -carotene.
Calculations for a fifty 1,000m2 Dunalliela salina farm suggest that the fixed
setup cost (site preparation, pond construction, production, and harvesting and
processing equipment) is between $1 and $1.5million. Using a five-year return-
of-investment period as a criterion to distribute the fixed cost, a conservative esti-
mate of the annual fixed cost per 1,000m2 would be $6,000 (0.21,500,000/50).
Annual variable cost (labor, CO2, nutrients, and electricity) is estimated to be
between $15,000 and $24,000 per 1,000m2. Using a very conservative approach,
total costs are estimated to be about $30,000 per 1,000m2. The most conservative
estimate of yield we have seen is 50kg of -carotene per 1,000m2. Thus, under
these conservative estimates, the break-even point is reached when a kilogram of
natural -carotene fetches $600per kg.
The estimates used thus far are quite conservative. Ben-Amotz and Avron (1980)
estimated that, with more experience and fine-tuning in production, annual yields
could rise to 120kg per 1,000m2 (based on harvesting 400t of Dunaliella salina on
50,000m2). It is assumed that algae have 30% dry matter, and 5% of it is -carotene.
Thus, -carotene per 1,000m2 is 0.05(400/500)(1,000)=120kg per 1,000m2. Thus,
using this yield estimate, assuming an annual variable cost of $15,000 per 1,000m2
and a fixed cost of $21,000 per 1,000m2, the investment in -carotene production can
be recaptured in 1year assuming a natural -carotene price of $300 per kg.
Phycobiliproteins are algal derivatives that have utility in diagnostic tools.
Specifically, biliproteins from microalgae (phycobiliproteins) are used as fluo-
rescent markers for genetic screening in cell analysis and imionchemialasi (Pulz
and Gross 2004). This application is based on a discovery made by Professor
Alexander Glazer of the University of California at Berkley and published in
1982. According to Professor Glazer, the first application started in 1983 when
two laboratories started producing phycobiliproteins. The case of phycobiliprotein,
like the case of agar, demonstrates the large range of commercial opportunities
algal products are starting to have, with sophistication and growth in biological
and genetic research, experimentation, and commercial application.
4.3.3Food and Feed Products
Many microalgae have a high nutritional value. They contain proteins, vitamins
and minerals, and non-saturated fats. Moreover, they can yield higher outputs for
the same levels of water and land. These characteristics led to the success of the
production of Spirolina, a microalgae that grown isolated in a monocultural set-
ting (another type of micro-algae grown isolated in a monoculture is Dunaliella
salina). It has commercial success as a health food and is a component of many
health food products. Worldwide, Spirulina is grown in many countries for animal
feeds and food nutrition supplement (FAO 2010). Other important species include
Chlorella, Dunaliella, Nostoc, and Aphanizomenon.
Culture of the freshwater algal Haematococcus pluvialis was developed in a
few countries and is used for the extraction of astaxanthin (FAO 2010), a natural
antioxidantsee also Del Campo et al. (2007). A price tag of $815 per kg has
been quoted in the literature (Benemann and Oswald 1996; Vonshak 1997; Lee
2001; Carlsson and Bowles 2007). Carlsson and Bowles (2007) suggest that cur-
rently the delivery price to the US from China for 20-ton containers is $5 per kg
for Spirulina and twice as high for Chlorella. Production costs have been esti-
mated at $25 US per kg.
Algae produced as a co-product of waste management plants may be used for
animal/livestock as well as other value added products (Lundquist et al. 2010;
Hochman et al. 2010). Much research is needed to develop large-scale food and
feed production from algaeone needed to identify species, production, pro-
cedures, etc. The food surplus problems in the United States and strong political
influence of grain farmers in America prevent production of feed from microalgae.

Note that, while there is no public support for research on obtaining food from
algae, there is much support for fuel production from algae. Growing lipid-rich
species of freshwater algae for biofuel production is the latest development in
freshwater algae. Compared with seaweed farming, the culture of freshwater algae
is generally poorly reported worldwide.
4.3.4Fatty Acids
Certain unsaturated fatty acids in triglycerides have desirable therapeutic and

health-promoting properties. Research has shown that omega-3 fatty acids reduce
cholesterol and fat levels in the blood and cleanse the lining of blood vessels
(Simopoulos 1991). The medical use of omega-3 fatty acids for prevention and
treatment of heart disease is increasing via prescribing fish oil to heart patients.
Usually, this treatment continues throughout the lifetime of the patients.
Moreover, some doctors prescribe similar dosages to individuals with high-risk
profiles with respect to coronary diseases. As evidence of the effectiveness of this
treatment spreads, its adoption is likely to grow. Studies (for example, Yetir 1988)
have shown that omega-3 fatty acids have the effective therapeutic properties dealing
with rheumatoid arthritis and immunodeficiency diseases, and doctors are consider-
ing prescribing pills derived from fish oils to combat these diseases. Cod and other
fish are not the direct producers of microalgae and extraction of omega-3 fatty acids
from the microalgae. The product extraction directly from the algae is likely to be
superior to the cod liver oil as (1) it will not have the off flavor of cod liver and (2) it
will be more pure product and thus more effective. The use of microalgae should
not be restricted to direct extraction of omega-3 fatty acids. They can also be used
as feed for chickens and dairy cows with will then introduce omega-3 fatty acids to
eggs and milk. The application to eggs may be especially useful since it will tend to
reverse (and combat) the contribution of eggs to cholesterol buildup.
The medical discoveries about the therapeutic properties of omega-3 fatty acids
suggest a very large market to algal-derived fatty acids. The superiority of the
microalgae derivative and the continued growth in demand for omega-3 fatty acids
suggest much higher sales potential for fatty acids derived from algae. There is
a substantial market for omega-3 fatty acids, and they can be marketed through
distributional channels of drugs and health products. Currently, species used to
produce fatty acids include Odontella aurita/Bacillariophyta, as well as Isochrysis
galbana/Chlorophyta and Phaedactylum Tricornutum/Bacillariophyta (Molina
Grima et al. 1994; Pulz and Gross 2004).
To assess the profitability of their production from microalgae, one has esti-
mates based on information available from other products. For the profitability
analysis, let P denote price per kilo of fatty acid, OC denote operational cost per
1,000m2, I denote investment per 1,000m2, and Y denote output of fatty acid
per 1,000m2. Output of fatty acid is the product of a share of fatty acid in dried
weight of algae (denoted by S) and production of dry matter of algae per 1,000m2
(denoted by Q), i.e., Y=Q * S.
Based on the price of cod liver oil pills, the retail price of omega-3 fatty acid is
estimated to be around $600 per kg. Next, it is assumed that producer price is only
20% of retail price (the other 80% covers transportation, processing, storage, and
marketing costs). Based on these conservative assumptions, the producer price we
use for profitability analysis is P=$120 per kg.
Based on several sources, 3t of dry eight of algae per acre is a low-end esti-
mate of annual production of most existing systems, and average output should be
about 6t per 1,000m2. (For brevity, the time dimension of yield and cost figures
are omitted, but all yield and cost figures given here are annual.) In the longer run,
with better knowledge and experience, dry weight algal production per 1,000m2
would reach 10t per 1,000m2. We use three levels of dry weight algal production:
Low output: Q=QL=3t per 1,000m2
Medium output: Q=QM=6t per 1,000m2
High output: Q=QH=10t per 1,000m2.
Experiments done (Koren et al. 1988) obtained results consistence with our
assumed levels. That team grew Nonochloropsis salina which is intended as a
source for fatty acid in a 100, a 2.5m2 pond, and a photoreactor. Production lev-
els are equivalent to annual yields per acre per year of between 3t of dry weight
(large pond) to 8t of dry weight (photoreactor).
Based on several sources, the share of fatty acids in dry weight of algae moved
from a low of 0.03 to high of 0.05. The experiments in Koren et al. (1988) resulted
in shares that are closer to 0.03. That makes us somewhat more cautious in the val-
ues we use for S in our assessment. The values we use are:
Low share: S=SL=0.03
High share: S=SH=0.04.
Combining the dry weight and share estimates, we obtain six values of yield
per acre, denoted for Y1 (lowest) to Y6 (highest). These values are:
Y1=low outputlow share=90kg/1,000m2/year
Y2=low outputhigh share=120kg/1,000m2/year
Y3=medium outputlow share=180kg/1,000m2/year
Y4=medium outputhigh share=240kg/1,000m2/year
Y5=high outputlow share=300kg/1,000m2/year
Y6=high outputhigh share=400kg/1,000m2/year.
As argued previously (in the case of -carotene), variable costs (operation cost)
per 1,000m2 are guesstimated to range from a low of OC = OCL =$15,000
per 1,000m2 to a high of OC = OCH =$24,000 per 1,000m2. Investment per
1,000m2 (which has no time dimension) is estimated to be I =$30,000 per
1,000m2. Then guesstimates have several implications.
Under the lowest yield assumption (Y = Y1) revenues cannot cover varia-
ble costs even when assuming low variable cost level. The deficit in this case is
$4,200 annually (15,00090120). The low variable costs are almost covered
even with the lowest algae production if the fatty acid in the dry algae is higher.
In the case of Y = Y2 and OC = OC2, the deficit is only $600 dollars annually
(15,00017,400).
When algae output is at a medium level the low level of variable cost is covered
and there is substantial surplus even assuming low fatty acid share. Specifically,
the surplus above variables cost when Y=Y3 and OC=OC3 is $6,600 annually
(18012015,000), and variable costs are at the high level, the annual deficit
is only $2,400 annually. Average algae output and higher fatty acid ratio (Y=Y4)
generated a surplus of $4,800 annually even when variable costs are at their high
level (4,8002401,20002,400).
When Algae output is in the high level, it leaves substantial surplus above
variable cost, even when share of fatty acids in dry weight is low. When Y=Y5
and OC = OCH, the annual surplus is $6,000 and fixed cost can be recovered
in 5years. When algae output is high and share us high (Y=Y6), surplus above
the high level of variable costs can be recovered in less than a year and a half.
Moreover, even when fatty price will decline by 50% (down to $60 per kg), rev-
enues will cover the variable costs.
4.3.5Polysaccharides
Polysaccharides are chemicals that are used as viscosifiers (thickening agents),

fluctuating agents, and lubricants. The value of polysaccharides varies according
to their use, availability, and purity. They include macroalgal derivatives such as
carrageenan and agar. Polysaccharides are derived from bacteria, fungi, and algae.
The bacteria and fungi are much more productive than algae, and genetic manipu-
lation and engineering of bacteria is in a much more advanced stage than genetic
engineering with algae. Still, algae generate complex and unique polysaccharides
and many algal derivatives are irreplaceable. Microalgae are the source of impor-
tant and commercially used polysaccharides, and the market for these algal deriva-
tives are in the hundred of millions of dollars.
Microalgae (such a Porphyridium cruentum/RhodophytaFuentes et al. 1999)
are commercially used to produce polysaccharides. Under the right conditions,
1555% of the weight of the microalgae can be extracted as polysaccharides.
Taking a very conservative approachassuming 15% polysaccharides share in
dry weight, medium yields (5t of dry weight algae per 1,000m2), and high cost
($30,000 per 1,000m2 annual total cost)the break-even price for polysaccha-
rides production is $40 per kg, which is within the medium range of market value
for polysaccharides. Taking a slightly more optimistic view30% polysaccha-
rides share in weight, medium yield (5t per 1,000m2), and low cost ($20,000
per 1,000m2)the break-even price is less than $15 per kg, quite a modest price
for many polysaccharides. Thus, polysaccharides from microalgae have good eco-
nomic potential.
4.3.6Food Coloring
There is a growing demand worldwide for organic food coloring. Regulating agen-
cies constantly limit the range of permissible chemical food coloring, and the regula-
tory process will be even stricter if and when organic substitutes are available. The
volume of the market for food color is immensein the billions of dollars annually.
Microalgae can be used as a source of many organic food coloring. As
Borowitzka and Borowitzka (1987) show, some microalgae contain substantial
amounts of other types of carotenes in addition to -carotene. Other types of color-
ing appear in microalgae as well. In pure form it can fetch up to $1,000 per kg. It
has been argued that the potential of microalgae, as a source of food coloring, is
limited because algal-derived food coloring is not photostable. Namely, they tend
to bleach with cooking. Nevertheless, in spite of this limitation, the potential mar-
ket for microalgae-derived food coloring is vast.
4.3.7Osmoregulators
These are carbohydrates that can affect osmotic processes. Glycerol is the most
notable member in this compound category, which included other commercially
viable products as well. Substantial weight of the dry weight of several algae,
e.g., up to 50% (Dunaliella salina), can be transformed to osmoregulators under
the appropriate conditions. Microalgae compete with bacteria and animal fat as
sources of osmoregulators. Research should and is likely to discover valuable
osmoregulators that can be produced from microalgae.
4.3.8Energy
The idea of using micro-algae to produce biodiesel is not a new idea and much
research has been allotted to the topic (Gallaghar 2011). A project at the National
Renewable Energy Laboratory has collected roughly 3,000 strains of algae from
northwest and southeast regions of the continent of the U.S. and Hawaii (Sheehan
et al. 1998). Carlsson and Bowles (2007) notes that the vlc-PUFAs may be less
appropriate for the production of biodiesel, since the polyunsaturation leads to oxi-
dation concerns in the fuel.
Gallaghar (2011) computed the Net-Present Value (NPV) using production and
cost figures reported in the literature. He concludes that high yield and high oil
prices, together with moderate government support and carbon prices, make bio-
diesel production viable economically. Assuming high yield (134mt/ha) with lipid
concentration of 40% leads to 6,430 gal/ac. Then, is we assume $1.00 subsidy
per gallon and a price of carbon of $44 per ton, a NPV of 17.4million dollars
is achieved with a payback period of 16.7 and an IRR of 12.4. Gallaghar (2011)
analysis suggests the once the environment and the social cost of pollution is intro-
duced into calculations, biodiesel production from algae can become economically
viable with moderate subsidies. See also work by Demirbas and Demirbas (2011).
4.3.9Other Applications
Microalgae contain many useful chemical compounds, and its derivatives can be
used in the future for many other applications in addition to the ones mentioned
above. They include cosmetic and skin products, food and feed supplements,
vitamins, and fertilizers, as well as fuel. Microalgae may be less productive than
bacteria, and our ability to manipulate it is much smaller. But microalgae con-
tains unique and complex products not available otherwise. Therefore, science
potentially can offer much research regarding the use and manipulation of micro-
algae. Today, in addition to the products discussed above, Spirulina is used to
produce phycocyanin and biomass (Lee 2001; Costa et al. 2003) while Chlorella
vulgaris/Chlorophyta is used to produce biomass (Lee 2001).
4.4Production Systems
4.4.1Open Ponds
We constructed some tentative estimates. First we present the following two esti-
mates of annual operational costs per 1,000m2,
High cost Low cost

Labor $10,000 $4,500
Feed (CO2 and nitrogen) 10,000 8,000
Energy, oil, and water 15,000 1,000
Machinery cost (short term) repair and maintenance 2,500 1,500
$24,000 $15,000
These operational costs include production, harvesting, and drying. The big dif-
ference between the high and low cost is in the labor cost estimates. Over time, as
experience is gained, work procedures will become better established and labor
costs will decline much further. Feed costs are the bug cost item in the long run,
especially the CO2 cost. The cost of repair and maintenance will decline with time
as more efficient production technologies and machinery are developed. Energy
efficiency is likely to increase over time, but energy consumption will rise as pro-
duction becomes more automated and labor intensive. As mentioned before, these
are very gross guesstimates; more knowledge on cost structures is required. Still,
all experts we discuss with agree that operational costs can be reduced further
$10,000 per 1,000m2 and that reducing CO2 and feed costs is major challenge.
When it comes to capital cost, it may be up to $30,000 per 1,000m2. Based on

501,000m2 ponds, up to $15,000 per 1,000m2 will be required for land, land prep-
aration, and pond construction. Equipment (piping, paddles, drying equipment etc.)
will require another $10,000 per acre, design management and coordination will
require the rest. Capital cost may be reduced over time to one third as experience
is added. Obviously, rise of yield and capital-intensive technologies (plastic testing)
may increase capital cost but, in this case, with substantial change in yields.
4.4.2Photobioreactors
These systems are different types of tanks or closed systems, in which algae is
grown (Richmond 2004). In these systems, water, nutrients and CO2 are supplied
in a controlled way, while oxygen is removed. Janssen et al. (2003) and Choi et al.
(2003) review developments in work that optimize photobioreactors systems for
algae cultivation.
4.4.3Heterotrophic
Others work suggests that algae can be grown in conventional fermentors instead
of photobioreactors to produce high value products (Wen and Chen 2003). Instead
of use of light and photosynthesis, heterotrophic utilize carbon sources in the
medium for the carbon and energy generation (Ward and Singh 2005).
4.5Summary and Conclusions
This chapter suggests that commercial utilization of algae, beyond biofuels, is eco-
nomically viable, and that there is a worldwide market for algal derivatives that
is estimated to be in the billion of US dollars. While application of algae as biofuel
has gained much attention, the literature suggests that some algal derivatives, which
researchers worked on during the last several decades, matured and proved quiet lucra-
tive. Others are still at the research and development stage, or are just been thought of.
Energy production from algae has gained much attention. Many algal species
are rich in oil content and algal is more productive at producing oil than any of
the existing terrestrial plants. While aquatic biomass may be used as raw material
for co-firing and producing electricity power and heat, much research and public
and private funds are channeled to the commercial development of the production
of bio-oil, biomethane, biodiesel or biogas. However, currently these technologies
cannot compete with fossil fuels. Policy can facilitate the adoption of these tech-
nologies of algae technologies as part of a green economy (Bangalore et al. 2012).
Such policy emphasis may misfire if the algae sector cannot stand on its own
feet and compete after a relatively short period of transition and learning. Hybrid
applications using algae to produce fuels and other product will increase the pro-
ductivity of biofuel sector and made bio-algae more efficient.
This chapter briefly surveyed the economic performance of the algae industry
thus far. It offers a snapshot of the algae industry and its potential. Further work,
however, is needed to better assess the economic viability of various algae deriva-
tives and to understand the true potential of algae and its impact on the energy sec-
tor. We leave this for future work.
Appendix A: Interviews
Abuoav J (Dr.,). Chief of Surgery, Mount Zion University, San Francisco, California.
Amit U. Ein Yahav, Arava, Israel.
Arad S (Dr.). Ben Gurion University, Box 1025, Beer Sheva, Israel, 84110.
Borowitzka M. School of Environmental and Life Sciences, Murdoch University,
Perth, Australia.
Ben-Amotz A (Professor). Israel Oceanographic Institute, Tel Shikmona, P.O. B.
8030, 31080 Haifa, Israel.
Foget RD. Manager Marketing Services, Bio Products, FMC Corporation, Marine
Colloids Division, 2000 Market Street, Philadelphia, PA, 19103.
Glazer AN (Professor). Department of Microbiology and Immunology, University of
California, Berkley, California, 94720.
Guron Y. B.A.R.D. Fund, P.O. Box Bet Dagan 50250, Israel.
Martinez W. USDA-ARS, Room 226, Building 005, BARC-West, Beltsville,
Maryland, 20705.
Neushul M (Professor). Marine Science Institute, University of California,
William J Department of Civil Engineering, University of California, Berkley,
California, 94720.
Oswald WJ Department of Civil Engineering, University of California, Berkley,
California, 94720.
Ramus J. Duke University Marine Lab., Beaufort, North Carolina, 28516.
Renn DW. (Dr.). FMC Corporation, 5 Maple Street, Rockland, Maine, 04841.
Sfat MR. Bio-Technical Resources Inc., 1035 South Seventh Street, Mainitowoc,
Wisconsin, 54220.
Vreeland V (Dr.). Department of Biology, University of California, Berkley,
California, 94720.
References
Bangalore M, Hochman G, Zilberman D (2012) Differences in the adoption of agricultural anaer-

obic digestion in Europe and the United States. Working paper
Ben-Amotz A, Avron M (1980) Glycerol, -carotene, and dry algae meal production by commer-
cial cultivation of Dunaliella. In: Shelef G, Soeded SJ (eds) Algae biomass. Elsevier North-
Holland Biomedical Press, Oxford, pp 603610
Benemann JR, Oswald WJ (1996) Systems and economic analysis of microalgae ponds for
conversion of CO2 to biomass. Final report (No. DOE/PC/93204--T5). Department of
Civil Engineering, Pitburgh Energy Technology Centre, US. Available at http://www.osti.
gov/energycitations/product.biblio.jsp?osti_id=493389
Borowitzka MA, Borowitzka LJ (1987) Vitamins and fine chemicals from micro-algae. In:
Borowitzka MA, Borowitzka LJ (eds) Micro-algal biotechnology. Cambridge University Press,
New York
Borowitzka LJ, Brown AD (1974) The salt relation of marine and halophilic species of
Dunalialla: the role of glycerol as a compatible solute. Arch Icrobiol 96:3752
Carlsson AS, Bowles DJ (2007) Micro-and macro-algae: utility for Industrial applications: out-
puts from the EPOBIO project, Sep 2007. CPL Press: Science Publisher, UK. (Report)
Chapman VJ (1970) Seaweed and the uses. Methuen and Company, London
Choi SL, Suh IS, Lee CG (2003) Lumostatic operation of bubble column photobioreactors for
Haematococcus pluvialis cultures using a specific light uptake rate as a control parameter.
Enzym Micro Technol 33:403409
Chynoweth DP (2002) Review of biomethane from marine biomass. In: Reith JH, Hal JW,
Lenstra WJ (eds) History, results and conclusions of the US Marine Biomass Energy
Program, (1968 to 1990), 194 pp
Costa JAV, Colla LM, Duarte P (2003) Spirulina platensis growth in open raceway ponds using
fresh water supplemented with carbon, nitrogen and metal ions. Zeits Naturforsch, C-A J
Biosci 58:7680
Del Campo JA, Garcia-Gonzales M, Guerrero MG (2007) Outdoor cultivation of microal-
gae for carotenoid production: current state and perspectives. Appl Microbiol Biotechnol
74:11631174
Demirbas A, Demirbas MF (2011) Importance of algae oil as a source of biodiesel. Energy
Conserv Mgmt 52:163170
FAO Fisheries and Aquaculture Secretariat (2010) The state of world fisheries and aquaculture
2010. Food and Agriculture Organization of the United Nation, Rome, Italy. Available at
http://41.215.122.106/dspace/handle/0/210
Fuentes MMR, Sanchez JLG, Sevilla JMF, Fernandez FGA, Perez JAS, Grima EM (1999)
Outdoor continuous culture of Porphyridium cruentum in a tubular photobioreactor: quantita-
tive analysis of the daily cyclic variation of culture parameters. J Biotechnol 70:271288
Gallagher Brian J (2011) The economics of producing biodiesel from algae. Renewable Energy
36(1):158162
Hochman G, Trachtenberg MC, Zilberman D (Forthcoming) Algae crops: co-production of algae
biofuels. In: Dierig D, Cruz VM (eds) Industrial crops: breeding for bioenergy and bioprod-
ucts. Springer Science, New York
Janssen M, Tramper J, Mur LR, Wijffels RH (2003) Enclosed outdoor photobioreactors: light
regime, photosynthetic efficiency, scale-up and future prospects. Biotechnol Bioeng 81:193210
Koren A, Amit U, Ilani T, Black B, Kahvvan A (1988) Research activities in the microalgae
growth in the Arava R&D center for aquaculture. Ein Yahav, Israel. (Report)
Lee YK (2001) Microalgal mass culture systems and methods: their limitation and potential. J
Appl Phycol 13:307315
Li Yanqun, Horsman Mark, Nan Wu, Lan Christopher Q, Dubois-Calero Nathalie (2008)
Biofuels from microalgae. Biotechnol Prog 24(4):815820
Lundquist TJ, Woertz IC, Quinn NWT, Benemann JR (2010) A realistic technology and engi-
neering assessment of algae biofuel production. Robert E. Kennedy Library, Cal Poly, San
Luis Obispo, CA. Available at http://digitalcommons.calpoly.edu/cenv_fac/188/
Luning K, Pang S (2003) Mass cultivation of seaweeds: current aspects and approaches. J Appl
Phycol 15:115119
McHugh DJ (2003) A guide to the seaweed industry. FAO fisheries technical paper no. 441,
Rome, FAO
Miao X, Wu Q, Yang CY (2004) Fast pyrolysis of microalgae to produce renewable fuels. J Anal
Appl Pyrol 71:855863
Molina Grima EM, Perez JAS, Camacho FG, Sevilla JMF, Fernandez FGA (1994) Effect of
growth-rate on the eicosapentaenoic acid and docosahexaenoic acid content of Isochrysis
galbana in chemostat culture. Appl Microbiol Biotechnol 41:2327
Munoz R, Guieysse B (2006) Algal-bacterial processes for the treatment of hazardous contami-
nants: a review. Water Res 40:27992815
Nishino H, Murakoshi M, Ii T, Takemura M, Kuchide M, Kanazawa M, Mou XY, Wada S,
Masuda M, Ohsaka Y, Yogosawa S, Satomi Y, Jinno K (2002) Carotenoids in cancer chemo-
prevention. Cancer Metastasis Rev 21(34):257264
Oswald WJ (1987a) Micro-algae and waste-water treatment. In: Borowitzka MA, Borowitzka LJ
(eds) Micro algal biotechnology. Cambridge University Press, New York, pp 305328
Oswald WJ (1987b) Large scale algal culture systems. In: Borowitzka MA, Borowitzka LJ (eds)
Micro algal biotechnology. Cambridge University Press, New York, pp 357394
Pulz O, Gross W (2004) Valuable products from biotechnology of microalgae. Appl Microbiol
Biotechnol 65:635648
Renn EW (1984) Agar and agarose: the indispensible partners in biotechnology. I&EC Res Devel
23:1721
Richmond A (2004) Principles for attaining maximal microalgal productivity in photobioreac-
tors: an overview. Hydrobiologia 512:3337
Sheehan John, Dunahay Terri, Benemann John, Roessler Paul (1998) A look back at the US
department of energys aquatic species program: biodiesel from algae, vol 328. National
Renewable Energy Laboratory, Golden
Shelef G (1982) High-rate algae ponds for waste water treatment and protein production. Water
Sci Technol 14:439452
Simopoulos AP (1991) Omega-3 fatty acids in health and disease and in growth and develop-
ment. Am J Clin Nutr 54(3):438463
Vonshak A (1997) Outdoor mass production of Spirulina: the basic concept. In: Vonshak A (ed)
Spirulina platensis (Arthrospira): physiology, cell-biology and biotechnology. Taylor &
Francis, London, pp 7999
Ward OP, Singh A (2005) Omega-3/6 fatty acids: alternative sources of production. Process
Biochem 40:36273652
Wen ZY, Chen F (2003) Heterotrophic production of eicosapentaenoic acid by microalgae.
Biotech Adv 21:273294
Wiley PE, Campbell JE, McKuin B (2011) Production of biodiesel and biogas from algae: a
review of process train options. Water Environ Res 83(4):326338
Yetir JZ (1988) Clinical application of fish oils. J Amer Med Assoc 260:665670
Part II
Biomass Biology
Chapter 5
Regional Gene Pools for Restoration,
Conservation, and Genetic Improvement
of Prairie Grasses
Andrew R. Jakubowski and Michael D. Casler
AbstractSwitchgrass (Panicum virgatum), big bluestem (Andropogon gerardii),

and Indiangrass (Sorghastrum nutans) are native warm-season grasses that have
been identified as potential cellulosic bioenergy feedstock crops due to their poten-
tial for high yields, perennial life habit, and nutrient use efficiency. This chap-
ter outlines the role that improved cultivars and unimproved locally collected
ecotypes can play in meeting agronomic and conservation goals. Improved cul-
tivars grown for use as a bioenergy feedstock will be established in areas where
introgression will occur with native populations. The concerns regarding the intro-
gression of transgenes or non-adaptive alleles are outlined along with several ave-
nues for mitigating these concerns. The agronomic and breeding history of each
species is reviewed, as well as their importance in the conservation and restoration
efforts of the prairie ecosystems of North America. We argue that both improved
and locally collected ecotypes can coexist on the landscape and help to jumpstart
the shift to a bioenergy based economy that provides sufficient biomass to meet
cellulosic bioenergy goals, restore native ecosystems, and provide an array of reg-
ulating, cultural, and supporting ecosystem services while increasing the sustain-
ability of agriculture.
A. R. Jakubowski(*)
Department of Agronomy, University of Wisconsin-Madison, 1575 Linden
Dr., Madison, Wisconsin 53706, USA
e-mail: Jakubowski@wisc.edu
M. D. Casler
USDA-ARS, U.S. Dairy Forage Research Center, 1925 Linden Dr., Madison,
Wisconsin 53706, USA
M. D. Casler
DOE-Great Lakes Bioenergy Research Center, University of Wisconsin-Madison,
445 Henry Mall, Madison, Wisconsin 53706, USA

68 A. R. Jakubowski and M. D. Casler
Keywords Big bluestem Cellulosic bioenergy Conservation biomass Cultivars

Indiangrass Introgression Local ecotypes Plant adaptation regions Plant
breeding Switchgrass
5.1Introduction
Perennial warm-season grasses have been identified as preferred cellulosic bioen-

ergy feedstock crops in temperate environments due to their high potential annual
yields and high water and nutrient use efficiency (Byrt et al. 2011). Their perennial
life habit eliminates the need for annual tillage and establishment, reducing estab-
lishment costs and greatly increasing their carbon sequestration potential (Adler
et al. 2007). In addition, their annual senescence cycle results in retranslocation
of aboveground nutrients to belowground organs, resulting in reduced nutrient
removal when plants are harvested following senescence (Clark 1977). While non-
native grasses have tremendous potential as a feedstock (Miscanthus x giganteus,
in particular), this chapter will focus on the role native grass species can play in
achieving agronomic and conservation goals related to bioenergy.
This chapter will focus on three native North American taxa, switchgrass
(Panicum virgatum L.), big bluestem (Andropogon gerardii Vitman), and
Indiangrass (Sorghastrum nutans (L.) Nash). These three species were the domi-
nant grasses of the North American tallgrass prairie, with native ranges extend-
ing from the Rocky Mountains to the Atlantic Ocean, north to central Canada
and south to the Gulf of Mexico (Fig.5.1). These three species were dominant
grasses in central North America, but today more than 80% of prairie has been
converted to other land uses (primarily agriculture), with tallgrass prairie reduced
by as much as 99% in some states or provinces (Samson and Knopf 1994). These
remnant prairie sites remain in unplowed areas, but have become increasingly
Indiangrass
Switchgrass
Big bluestem
Fig.5.1Map of estimated native range of switchgrass, big bluestem, and Indiangrass in Canada
and the USA. Data adapted from Barkworth et al. (2003)
5 Regional Gene Pools for Restoration, Conservation 69
isolated from one another (Simberloff and Gotelli 1984). To date, the vast major-
ity of planting of these species has been for prairie restoration. While these efforts
have grown exponentially over the past 20years, they have reclaimed only a frac-
tion of the land once in prairie (Anderson 2009). The use of native warm-season
grasses for use as a bioenergy feedstock will greatly expand the presence of these
species on the landscape. We argue that the planting and breeding of these three
species, with an understanding of their geographic and genetic context, can meet
agronomic and conservation goals while making these species common once again
in North America.
5.2Species Background
5.2.1Switchgrass
Considerable interest has been shown in using switchgrass as cellulosic bioenergy

feedstock crop. Switchgrass was identified in part because of its favorable traits
and in part due to serendipity (Parrish et al. 2012). Of the three species discussed
within this chapter, switchgrass has been the focus of the majority of research
efforts. There are two primary taxonomic groups of switchgrass, upland and low-
land ecotypes. Upland ecotypes are adapted to northern regions and consist of
primarily two cytotypes. Upland octoploids (2n=8x=72) are the most com-
mon ploidy for wild and cultivated accessions, while tetraploids (2n=4x=36)
are slightly less frequent (Zhang et al. 2011b). Plants ranging from 2x to 12x
have been identified, but these cytotypes are rare (Costich et al. 2010). Lowland
ecotypes are adapted to southern regions, are later flowering, and generally more
productive than upland ecotypes. Lowland ecotypes are predominantly tetraploids,
but several octoploid accessions have recently been confirmed (Zalapa et al. 2011;
Zhang et al. 2011a). The center of diversity is in the Gulf Coast region and anal-
ysis of population structure suggests relatively large interbreeding populations
occurring at the scale of hundreds of kilometers (Zhang et al. 2011b).
Yields of switchgrass vary widely depending upon the region and the ecotype
grown. A meta-analysis of 39 field trials found an average yield of upland
ecotypes of 8.74.2 and 12.95.9Mg/ha in lowland ecotypes (Wullschleger
et al. 2010).
5.2.2Big Bluestem
Big bluestem has drawn interest as a potential biofuel crop recently (Anderson
et al. 2008; Zhang et al. 2012). The species is primarily found as a hexaploid
(2n=6x=60) with enneaploid (2n=9x=90) less common (Norrmann et al. 1997).
Hybridization between the two cytotypes is possible and aneuploids appear to be com-
mon (Keeler 1992). Established plants grown in agronomic settings have been iden-
tified as tall as 3m, suggesting there is great potential for producing high yielding
stands of bluestem (Stubbendieck et al. 1991). The species has also been shown to
have higher fermentability potential when compared to other warm-season grass spe-
cies (Jung and Vogel 1992; Weimer and Springer 2007). In agriculturally managed
systems, big bluestem was found to produce between 9.9 and 15.7Mg/ha of biomass
depending on fertilization in Iowa, USA (Hall et al. 1982). In an evaluation of produc-
tivity in unmanaged systems in the Great Plains, big bluestem was found to produce
more than twice as much biomass as either switchgrass or Indiangrass (Epstein et al.
1998). Big bluestem was also reported to have produced twice as much biomass per
unit of applied nitrogen than switchgrass and Indiangrass (Perry and Baltensperger
1979). Recent work found three overlapping gene pools in the species from the Great
Plains, Midwest, and Northeast USA, suggesting large interbreeding populations simi-
lar to those found in switchgrass (Price et al. 2012).
5.2.3Indiangrass
Indiangrass has seen considerably less interest as a bioenergy feedstock.

However, there has been research into its productivity and the heritability of traits
for use as a forage crop. Three cytotypes have been identified (2n=2x=20,
2n =4x=40, and 2n=6x=60) with the tetraploid cytotype most common
(Gould 1975; Riley and Vogel 1982). Biomass yields of three newly released
Indiangrass varieties (Chief, Scout, and Warrior) averaged 10.2Mg/ha when
grown in Nebraska, USA (Vogel et al. 2010). Heritability estimates for yield and
in vitro dry matter digestibility (IVDMD) were calculated to be 0.43 and 0.42,
respectively (Vogel et al. 1981). There have been no large-scale evaluations of the
population structure of the species using modern genetic markers, but the simi-
larity in geographic distribution and reproductive traits with switchgrass and big
bluestem suggests there may be a similar population structure in Indiangrass. A
comparison of switchgrass, Indiangrass, and big bluestem in Iowa, USA found
Indiangrass and switchgrass sward plots to have similar productivities, while big
bluestem produced significantly less (Wilsey 2010). Additional research is neces-
sary to evaluate germplasm to determine the yield potential of Indiangrass outside
of the Great Plains, USA.
5.3Breeding
There are many potential benefits to using perennial warm-season grasses for
biomass production, but all three species have similar characteristics in need of
improvement. Establishment of all crops is slow and yields are a fraction of their
potential yields during the establishment season (Parrish and Fike 2005). The
seeds of all three species maintain dormancy and require up to a year of storage
to achieve optimal germination (Beckman et al. 1993; Emal and Conard 1973).
These species also emerge later in the spring than cool-season plants, creating
potential weed pressure and an incomplete use of the growing season (Sanderson
et al. 2004). All three species have seed shattering habits that will reduce the effi-
ciency of breeding and large-scale seed production. However, sustained breeding
efforts should be capable of improving all of these traits.
Previous breeding efforts in these species suggest that there is potential for
improvements in yield and quality characteristics, with the majority of work
occurring in switchgrass. Yields are expected to increase by 50100% through a
combination of conventional breeding, genetic modification, and the development
of hybrid switchgrass (crosses between upland and lowland ecotypes) in the next
1020years (Casler 2010; Martinez-Reyna and Vogel 2008; Vogel and Mitchell
2008). Evaluations of genetic variability for favorable bioenergy traits appear to
be high in switchgrass (Das et al. 2004; Rose et al. 2008). The early recognition
of these three species as potential forage crops has resulted in significant col-
lection of germplasm for breeding that is publically available within the USDA
Germplasm Resources Information Network (GRIN) system. As of November
2012, there are 332, 1,188, and 54 accessions in GRIN for switchgrass, big
bluestem, and Indiangrass, respectively. Additional collections of the three species
from remnant prairies are underway throughout the USA.
This tremendous increase in potential productivity can make the economics of
growing perennial grasses for bioenergy much more favorable (Perrin et al. 2008).
Previous breeding efforts in native warm-season grasses have proven successful,
although programs have focused on improving forage quality and nutritional value
for animal grazing until recent efforts in switchgrass for bioenergy (Anderson
etal. 2008; Mitchell et al. 2005; Vogel et al. 2010). Cultivars of all three spe-
cies have been released with improved forage quality using phenotypic selection
of IVDMD (Hopkins et al. 1993; Mitchell et al. 2005; Vogel et al. 2006; Vogel
et al. 2010). These cultivars also show minor improvements in yield. Only switch-
grass has had improved varieties specifically for use as a bioenergy feedstock crop
released by public breeders (Burns et al. 2010; Wu and Taliaferro 2009) and at
least one private company (Ceres, Inc Thousand Oaks, CA).
The use of genetic modification to improve plants for use as a bioenergy feed-
stock has been called imperative for making bioenergy crops viable (Gressel
2008). Indeed, the down-regulation of genes involved in lignification improved
ethanol yield in switchgrass by up to 38% with significantly reduced pretreat-
ment requirements and conversion costs (Fu et al. 2011). However, field testing of
these varieties is required to confirm their improvement as conventional breeding
for lower lignin in switchgrass and other perennial crops has been shown to dra-
matically reduce plant fitness (Casler 1997; Casler et al. 2002). Whether herbicide-
resistance transgenes will be incorporated into publically available varieties in the
future is unknown; however, herbicide resistant lines of switchgrass have been
developed (Song et al. 2012).
5.4Gene Flow Concerns and the Role of Unimproved

and Local Varieties
While breeding efforts will be important for improving yield and quality in
these species, there are significant concerns regarding the breeding and introduc-
tion of improved native perennial crops (Kwit and Stewart 2012; Lonsdale and
FitzGibbon 2011). Because these species are native throughout North America,
there is fear that improved or introduced populations will introgress with native
populations and result in reduced fitness of wild populations or the develop-
ment of novel invasive populations (Lesica and Allendorf 1999; Selbo and Snow
2005; Byrne and Stone 2011). Many of the traits identified as targets of breeding
programs, such as rapid growth rates, plasticity across a range of environments,
high yield, and cold and drought tolerance, have been identified as traits asso-
ciated with increased invasion potential (Heaton et al. 2008; van Kleunen et al.
2010; Mclaughlin et al. 1999; Raghu et al. 2006; Sakai et al. 2001; Theoharides
and Dukes 2007). These species also have weedy relatives with which hybridiza-
tion may be possible (Ahrens et al. 2011). If genetically modified varieties of these
species are released (particularly with herbicide resistance), this concern of intro-
gression into weedy relatives is magnified further (Bagavathiannan et al. 2010;
Stewart et al. 2003; Warwick et al. 2009).
Much work remains in evaluating the pollen flow dynamics of these species.
The closest analog to pollen flow studies was undertaken with genetically mod-
ified creeping bentgrass (Agrostis stolonifera, Watrud et al. 2004; Zapiola et al.
2008). This species is a wind-pollinated perennial that is highly outcrossing and
has a synchronous flowering period with weedy relatives with which hybridiza-
tion is possible (Belanger et al. 2003). Systematic dispersal studies using senti-
nel plants recovered herbicide resistance seedlings at a distance of up to 21km
(Watrud et al. 2004). A follow up study collected plants from a 4.8km area sur-
rounding the initial planting area identified nine transgenic plants out of 20,400
tested (Zapiola et al. 2007). Switchgrass, Indiangrass, and big bluestem are likely
to disperse pollen over longer distances than creeping bentgrass due to their taller
morphology and higher pollen production (Kausch et al. 2010). Recent work has
shown that switchgrass pollen can remain viable under ideal field conditions for
up to 2.5h (Ge et al. 2011).
There is little previous knowledge on which to assess the risks of creating
novel invasive species through conventional breeding. One example used reed
canarygrass (Phalaris arundinacea L.) as a model for risk assessment of breed-
ing native perennial crops (Jakubowski et al. 2011). Reed canarygrass is native
to North America and Eurasia, but is considered a noxious wetland invader
throughout much of North America. The species is also planted as a forage
crop and improved varieties have been released in the North American market-
place. Cultivation had been suggested as the origin of the development of inva-
siveness in the species (Merigliano and Lesica 1998). However, Jakubowski
et al. (2011) showed that cultivars were more productive and fecund than wild
Switchgrass breeding
program
Fig.5.2Plant adaptation regions (PARs) to determine regional gene pools based on climatic and
geographic data. The locations of the 12 public switchgrass breeding programs are identified by
the open circles. Map developed using the methods of Vogel et al. (2005)
Eurasian populations in the environments for which they were selected (uplands
with nitrogen addition), but no more productive in the environment in which they
are considered invasive (wetlands). The authors conclude that breeding per se is
not responsible for invasion by the species, but the widespread planting of reed
canarygrass for forage and soil stabilization made the species common on the
landscape.
These concerns, along with the growing chorus for the recognition of the addi-
tional ecosystem services beyond production provided by the use of perennial
grasses in agriculture, encourage the use of locally collected ecotypes of native
grasses (Boody et al. 2005; Gasparatos et al. 2011). Locally collected seed is
defined as unimproved germplasm collected from remnant sites within a speci-
fied distance of the planting site. The geographic scale at which seed is considered
local varies by species (McKay et al. 2005), but should be determined by genetic
analysis as the scale at which interbreeding populations of a species exist. Vogel
et al. (2005) developed general guidelines for determining the geographic scale of
local germplasm using climatic and geographical variables by combining Baileys
Ecoregions with the USDA Plant Hardiness Zone maps. These regions were named
Plant Adaptation Regions (PARs) and were developed to guide collection and
selection of germplasm for natural area restoration when resources are unavailable
for genetic analysis (Fig.5.2). For the obligate out-crossing and wind-pollinated
grasses discussed here, local seed would be collected within approximately 500km
of a site. As mentioned earlier, up to 99% of land originally in tallgrass prairie
has been converted to other land uses. The planting of locally collected varieties
of these species would be an economically valuable way of maintaining genetic
resources on the landscape. Most cultivars and local collections are widely adapted
within one hardiness zone north and south of their origin (Casler 2012), which
includes multiple plant adaptation regions. Thus, the concept of local can be
quite broad, covering a broad geographic area for many cultivars. Public breed-
ing programs can play an important role in conserving germplasm and releasing
locally adapted varieties. The 12 programs involved in switchgrass breeding are
scattered throughout the native range of the species and test populations in many of
the PARs (Fig.5.2).
Further, collection and use of locally collected varieties may prove to be more
productive than improved varieties from distant locations. Numerous studies com-
paring production of local and non-local varieties in warm-season grasses have
found that locally collected varieties are more productive than non-local ecotypes
(Casler 2005; Casler et al. 2004; Gustafson et al. 2004; McMillan 1964, 1965).
The breeding of perennial grasses is a slow process that requires many years of
selection and evaluation before an improved variety is released (Casler 2010;
Yamada et al. 2005). This process is expected to accelerate with the increasing
availability of genomic tools, but is still limited by the generation time of peren-
nial crops. This process is further complicated by genotype by environment inter-
actions and the wide range of marginal lands upon which perennial grasses will
be established. The development of improved varieties in which improvements
are stable across a wide range of environments has proven difficult (Eberhart
and Russell 1966; Simmonds 1991). A comparison of productivity and quality
of locally collected wild accessions and cultivars (developed in Nebraska) of big
bluestem found no consistent differences in production or grazing tolerance in pas-
tures in Wisconsin, USA (Chamberlain et al. 2012). Emerging work in understand-
ing interactions between plant and mychorrizal fungi provenance suggests that
mismatched plant and fungal communities may have major impacts on plant pro-
duction, particularly in nutrient limited environments (Collins-Johnson et al. 2010,
Klironomos 2003). A home-field advantage has been identified in a range of
plant species when grown in the same soils from which they were collected (Smith
et al. 2012; Whitham et al. 2012).
Increasingly, the use of biomass harvest in conservation and restoration has
been identified as a sound management tool (Jakubowski et al. 2010; Gasparatos
etal. 2011; Hull et al. 2011). Land managers often have limited resources to con-
duct intensive management, such as the use of burning and herbicide to control
undesirable species. Partnerships between land managers and local farmers to har-
vest biomass from these lands historically considered closed to agricultural pro-
duction are becoming more common. The farmer harvests biomass and is allowed
to retain any benefits gained from the sale or processing of biomass, while the land
manager is receiving weed control and nutrient management at no cost. The value
of the biomass serves as an economic incentive to encourage restoration and con-
servation for protection of soil, water, and faunal resources. The development of
best management practices with a concrete basis in conservation has been devel-
oped to assure appropriate use of these lands (Ventura et al. 2012). Because these
lands are not managed for maximum agricultural production, the use of the highest
yielding variety is not necessary, nor appropriate. In addition, these types of lands
will be important in providing cellulosic biomass without establishment costs dur-
ing the early phases of a shift to a bioenergy economy.
5.5Coexistence of Improved Varieties and Conservation

Priorities
There is room for the coexistence of improved varieties (including genetic modi-
fication) of native warm-season grasses and native populations. The development
of improved cultivars with significant gains in yield and ease of establishment will
be critical in realizing the economic and ecological benefits of harvesting peren-
nial biomass crops for bioenergy. Yield gains achieved through breeding may prove
to be the most important factor in making bioenergy crops economically feasible.
However, concerns regarding development of novel invasive species and gene flow
from improved or GM varieties into wild populations are well founded. Thoughtful
policy and strategies are required to minimize these risks (Kausch et al. 2010).
Several agronomic and breeding strategies can also minimize these risks.
Breeding often selects improvements for a very specific target population of envi-
ronments, and stability across a wide range of environments has proved elusive for
even the most highly bred crops (Simmonds 1991). Selecting cultivars for specific
environments or requiring specific management (i.e. high nitrogen environments
or cultivation) reduces the probability that a crop will escape cultivation. Breeders
generally strive for cultivar stability across a range of environments, as these culti-
vars are more marketable and simplify selection for a producer. In contrast to typi-
cal cultivar development, a range of cultivars that are each designed for a specific
environment may be preferable for bioenergy feedstock crops. In addition, the
evaluation of crops in environments in which they have the potential to be inva-
sive may be appropriate to select against the populations that are highly productive
in these environments. This will require collaboration between breeders and weed
scientists to determine which traits and environments are most important.
Another safeguard to reduce the chance of creating a novel invader is to select
for reduced fecundity of perennial crops. Efforts to select for slowed maturity and
reduced flowering to improve forage quality have been successful in several per-
ennial pasture species (Buxton 1996; Casler et al. 2004). In addition, research to
reduce or eliminate flowering by altering the genetic mechanisms involved in the
flowering pathway has shown promise (Salehi et al. 2005; Jung and Mller 2009).
This technique not only reduces the fecundity of the crop but can also increase
biomass production due to a reallocation of resources. Selecting for reduced
fecundity in biomass crops has the potential to reduce the risks of introgression
of improved genes or traits into native populations and the escape of improved
varieties into undesirable environments. However, this will reduce the ability to
produce seed of new cultivars. Alternatively, the development of winter hardy
lowland varieties of switchgrass is underway due to their significantly higher

yields over upland varieties when grown at Northern latitudes (Casler 2012).
Lowland varieties grown at Northern latitudes have a 46week delay in flower-
ing time compared to upland varieties. This de facto reproductive barrier should
reduce the introgression of improved genes or traits into upland native popula-
tions in Northern latitudes, but not in Southern latitudes where native populations
of lowland plants exist. The development of male-sterility systems would compli-
cate breeding efforts, but would eliminate pollen dispersal and reduce the risk of
transgene escape (Kausch et al. 2010).
5.6Conclusion
The use of both improved and locally collected varieties can coexist on the land-
scape and help to jumpstart the shift to a bioenergy based economy that provides
sufficient biomass to meet cellulosic bioenergy goals, restore native ecosystems,
and provide an array of regulating, cultural, and supporting ecosystem services
while increasing the sustainability of agriculture. Breeders should appreciate the
concerns of the conservation community regarding the introgression of exotic
germplasm into native populations and incorporate appropriate safeguards to
reduce these risks into their breeding programs. Conservationists should appreci-
ate the importance of breeding in enabling a shift to a bioenergy economy and the
multitude of ecosystem services associated with such a shift. Most importantly, the
communities should work together to develop strategies that appropriately utilize
germplasm, protect natural ecosystems, and meet agronomic goals.
References
Adler PR, Del Grosso SJ, Parton WJ (2007) Life-cycle assessment of net greenhouse-gas flux for
bioenergy cropping systems. Ecol Appl 17:675691
Ahrens C, Ecker G, Auer C (2011) The intersection of ecological risk assessment and plant com-
munities: an analysis of Agrostis and Panicum species in the northeastern U.S. Plant Ecol
212:16291642
Anderson R (2009) History and progress of ecological restoration in tallgrass prairie. In:
Taylor C, Taft J, Warwick C (eds) INHS special publication 30: canaries in the catbird seat.
University of Illinois, Champaign-Urbana, pp 217228
Anderson W, Casler MD, Baldwin B (2008) Improvement of perennial forage species as feed-
stock for bioenergy. In: Vermerris W (ed) Genetic improvement of bioenergy crops. Springer,
New York, pp 309346
Bagavathiannan M, Spk A, Van Acker R (2010) Commercialization of perennial GE crops:
looming challenges for regulatory frameworks. J Agric Environ Ethics 24:227242
Barkworth M, Capels K, Long S, Piep M (2003) Flora of North America, vol. 25.
Magnoliophyta: Commelinidae (in part). http://herbarium.usu.edu/webmanual
Beckman JJ, Moser LE, Kubik K, Waller SS (1993) Big bluestem and switchgrass establishment
as influenced by seed priming. Agron J 85:199202
Belanger FC, Meagher TR, Day PR et al (2003) Interspecific hybridization between Agrostis
stolonifera and related Agrostis species under field conditions. Crop Sci 43:240246
Boody G, Vondracek B, Andow DA et al (2005) Multifunctional agriculture in the United States.
Bioscience 55:2738
Burns JC, Godshalk EB, Timothy DH (2010) Registration of colony lowland switchgrass. J
Plant Registr 4:189194
Buxton DR (1996) Quality-related characteristics of forages as influenced by plant environment
and agronomic factors. Anim Feed Sci Technol 59:3749.
Byrne M, Stone L (2011) The need for duty of care when introducing new crops for sustain-
able agriculture. Curr Opin Environ Sustain 3:5054
Byrt CS, Grof CPL, Furbank RT (2011) C4 plants as biofuel feedstocks: optimising biomass
production and feedstock quality from a lignocellulosic perspective. J Integr Plant Biol
53:120135
Casler MD (1997) Breeding for improved forage quality: potentials and problems. In:
Proceedings of the 18th international grassland congress. Winnipeg, Manitoba and
Saskatoon, Saskatchewan (Canada), pp 323330
Casler MD (2005) Ecotypic variation among switchgrass populations from the northern USA.
Crop Sci 398:388398
Casler MD (2010) Changes in mean and genetic variance during two cycles of within-family
selection in switchgrass. Bioenergy Res 3:4754
Casler MD (2012) Switchgrass breeding, genetics, and genomics. In: Monti A (ed) Switchgrass:
a valuable biomass crop for energy. Springer, London, pp 2954
Casler MD, Buxton DR, Vogel KP (2002) Genetic modification of lignin concentration affects
fitness of perennial herbaceous plants. Theor Appl Gen 104:127131
Casler MD, Vogel K, Taliaferro C (2004) Latitudinal adaptation of switchgrass populations. Crop
Sci 44:293303
Chamberlain SK, Paine LK, Harrison JL, Jackson RD (2012) Tradeoffs in performance of native
warm-season grass cultivars and locally harvested seed managed for wildlife habitat or live-
stock production. Agron J 104:1383
Clark FE (1977) Internal cycling of nitrogen in shortgrass prairie. Ecology 58:13221333
Collins Johnson N, Wilson GWT, Bowker MA et al (2010) Resource limitation is a driver of
local adaptation in mycorrhizal symbioses. Proc Natl Acad Sci USA 107:20932098
Costich DE, Friebe B, Sheehan MJ et al (2010) Genome-size variation in switchgrass (Panicum
virgatum): flow cytometry and cytology reveal rampant aneuploidy. Plant Genome 3:130141
Das MK, Fuentes RG, Taliaferro CM (2004) Genetic variability and trait relationships in switch-
grass. Crop Sci 44:443448
Eberhart SA, Russell WA (1966) Stability parameters for comparing varieties. Crop Sci 6:3640
Emal J, Conard E (1973) Seed dormancy and germination in Indiangrass as affected by light,
chilling, and certain chemical treatments. Agron J 65:383385
Epstein H, Lauenroth W, Burke I, Coffin D (1998) Regional productivities of plant species in the
Great Plains of the United States. Plant Ecol 134:173195
Fu C, Mielenz JR, Xiao X et al (2011) Genetic manipulation of lignin reduces recalcitrance and
improves ethanol production from switchgrass. Proc Natl Acad Sci USA 108:38033808
Gasparatos A, Stromberg P, Takeuchi K (2011) Biofuels, ecosystem services and human wellbe-
ing: Putting biofuels in the ecosystem services narrative. Agric Ecosys Environ 142:111128
Ge Y, Fu C, Bhandari H et al (2011) Pollen viability and longevity of switchgrass. Crop Sci
51:26982705
Gould FW (1975) The grasses of texas. Texas A&M University Press, College Station, TX
Gressel J (2008) Transgenics are imperative for biofuel crops. Plant Sci 174:246263
Gustafson DJ, Gibson DJ, Nickrent DL (2004) Conservation genetics of two co-dominant grass
species in an endangered grassland ecosystem. J Appl Ecol 41:389397
Hall K, George J, Riedl R (1982) Herbage dry matter yields of switchgrass, big bluestem, and
indiangrass with N fertilization. Agron J 74:4751
Heaton EA, Flavell RB, Mascia PN et al (2008) Herbaceous energy crop development: recent
progress and future prospects. Curr Opin Biotechnol 19:202209
Hopkins AA, Vogel KP, Moore KJ (1993) Predicted and realized gains from selection for in vitro
dry matter digestibility and forage yield in switchgrass. Crop Sci 33:253258
Hull S, Arntzen J, Bleser C et al (2011) Wisconsin sustainable planting and harvest guidelines for
nonforest biomass. Madison, WI
Jakubowski AR, Casler MD, Jackson RD (2010) The benefits of harvesting wetland invaders for
cellulosic biofuel: an ecosystem services perspective. Restor Ecol 18:789795
Jakubowski AR, Casler MD, Jackson RD (2011) Has selection for improved agronomic traits
made reed canarygrass invasive? PLoS ONE 6:e25757
Jung C, Mller AE (2009) Flowering time control and applications in plant breeding. Trends in
Plant Science 14:563573
Jung H, Vogel K (1992) Lignification of switchgrass (Panicum virgatum) and big bluestem
(Andropogon gerardii) plant parts during maturation and its effect on fibre degradability. J
Sci Food Agric 59:169176
Kausch A, Hague J, Oliver M, Li Y (2010) Transgenic perennial biofuel feedstocks and strategies
for bioconfinement. Biofuels 1:163176
Keeler KH (1992) Local polyploid variation in the native prairie grass Andropogon gerardii.
Amer J Bot 79:12291232
Klironomos J (2003) Variation in plant response to native and exotic arbuscular mycorrhizal
fungi. Ecology 84:22922301
Kwit C, Stewart C (2012) Gene flow matters in switchgrass (Panicum virgatum L.), a potential
widespread biofuel feedstock. Ecol Appl 22:37
Lesica P, Allendorf FW (1999) Ecological genetics and the restoration of plant communities: mix
or match? Restor Ecol 7:4250
Lonsdale W, FitzGibbon F (2011) The known unknownsmanaging the invasion risk from bio-
fuels. Curr Opin Environ Sustain 3:3135
Martinez-Reyna JM, Vogel KP (2008) Heterosis in switchgrass: spaced plants. Crop Sci
48:13121320
McKay JK, Christian CE, Harrison S, Rice KJ (2005) How local is local? A review of practical
and conceptual issues in the genetics of restoration. Restor Ecol 13:432440
Mclaughlin S, Bouton J, Bransby D, et al. (1999) Developing switchgrass as a new crop. In:
Janick J (ed) Proceedings fourth national new crops symposium. ASHS Press, Alexandria,
VA, pp 282299
McMillan C (1964) Ecotypid differentiation within four North American prairie grasses. I.
Morphological variation within transplanted community fractions. Amer J Bot 51:11191128
McMillan C (1965) Ecotypic differentiation within four North American prairie grasses. II.
Behavioral variation within transplanted community fractions. Amer J Bot 52:5565
Merigliano MF, Lesica P (1998) The native status of reed canarygrass (Phalaris arundinacea L.)
in the Inland Northwest. USA. Nat Areas J 18:223230
Mitchell RB, Vogel KP, Klopfenstein TJ et al (2005) Grazing evaluation of big bluestems bred
for improved forage yield and digestibility. Crop Sci 45:22882292
Norrmann G, Quarin C, Keeler K (1997) Evolutionary implications of meiotic chromosome
behavior, reproductive biology, and hybridization in 6x and 9x cytotypes of Andropogon
gerardii (Poaceae). Amer J Bot 84:201207
Parrish DJ, Fike JH (2005) The biology and agronomy of switchgrass for biofuels. Crit Rev Plant
Sci 24:423459
Parrish DJ, Casler MD, Monti A (2012) The evolution of switchgrass as an energy crop. In:
Monti A (ed) Switchgrass. Springer, London, pp 128
Perrin R, Vogel K, Schmer M, Mitchell R (2008) Farm-scale production cost of switchgrass for
biomass. Bioenergy Res 1:9197
Perry L, Baltensperger D (1979) Leaf and stem yields and forage quality of three N-fertilized
warm-season grasses. Agron J 71:355358
Price DL, Salon PR, Casler MD (2012) Big bluestem gene pools in the Central and Northeastern
United States. Crop Sci 52:189200
Raghu S, Anderson RC, Daehler CC et al (2006) Adding biofuels to the invasive species fire.
Science 313:1742
Riley R, Vogel KKP (1982) Chromosome numbers of released cultivars of switchgrass, indian-
grass, big bluestem, and sand bluestem. Crop Sci 22:10821083
Rose LW, Das MK, Taliaferro CM (2008) Estimation of genetic variability and heritability for
biofuel feedstock yield in several populations of switchgrass. Ann Appl Biol 152:1117
Sakai AK, Allendorf FW, Holt JS et al (2001) The population biology of invasive species. Annu
Rev Ecol System 32:305332
Salehi H, Ransom CB, Oraby HF, et al. (2005) Delay in flowering and increase in biomass of
transgenic tobacco expressing the Arabidopsis floral repressor gene FLOWERING LOCUS
C. J Plant Physiol 162:711717
Samson F, Knopf F (1994) Prairie conservation in North America. Bioscience 44:418421
Sanderson MA, Schnabel RR, Curran WS et al (2004) Switchgrass and big bluestem hay, bio-
mass, and seed yield response to fire and glyphosate treatment. Agron J 96:16881692
Selbo SM, Snow AA (2005) Flowering phenology and genetic similarity among local and
recently introduced populations of Andropogon gerardii in Ohio. Restor Ecol 13:441447
Simberloff D, Gotelli N (1984) Effects of insularisation on plant species richness in the prairie-
forest ecotone. Biol Conserv 29:2746
Simmonds NW (1991) Selection for local adaptation in a plant-breeding program. Theor Appl
Genet 82:363367
Smith DS, Schweitzer JA, Turk P et al (2012) Soil-mediated local adaptation alters seedling sur-
vival and performance. Plant Soil 352:243251
Song G, Walworth A, Hancock JF (2012) Factors influencing Agrobacterium-mediated transfor-
mation of switchgrass cultivars. Plant Cell, Tiss Org Cult 108:445453
Stewart CN, Halfhill MD, Warwick SI (2003) Transgene introgression from genetically modified
crops to their wild relatives. Nat Rev Genet 4:806817
Stubbendieck J, Hatch S, Butterfield C (1991) North American range plants. University of
Nebraska Press, Lincoln, NE
Theoharides KA, Dukes JS (2007) Plant invasion across space and time: Factors affecting nonin-
digenous species success during four stages of invasion. New Phytol 176:256273
Van Kleunen M, Weber E, Fischer M (2010) A meta-analysis of trait differences between inva-
sive and non-invasive plant species. Ecol Lett 13:235245
Ventura S, Hull S, Jackson R et al (2012) Guidelines for sustainable planting and harvest of non-
forest biomass in Wisconsin. J Soil Water Conserv 67:17A20A
Vogel KP, Mitchell RB (2008) Heterosis in switchgrass: biomass yield in swards. Crop Sci
48:21592164
Vogel K, Gorz H, Haskins F (1981) Heritability estimates for forage yield, in vitro dry matter
digestibility, crude protein, and heading date in indiangrass. Crop Sci 21:3538
Vogel KP, Schmer MR, Mitchell RB (2005) Plant adaptation regions: ecological and climatic
classification of plant materials. Range Ecol Man 58:315319
Vogel KP, Mitchell RB, Klopfenstein TJ, Anderson BE (2006) Registration of Goldmine big
bluestem. Crop Sci 46:23142315
Vogel KP, Mitchell RB, Gorz HJ et al (2010) Registration of Warrior, Scout, and Chief
Indiangrass. J Plant Registr 4:115
Warwick SI, Beckie HJ, Hall LM (2009) Gene flow, invasiveness, and ecological impact of
genetically modified crops. Ann New York Acad Sci 1168:7299
Watrud LS, Lee EH, Fairbrother A et al (2004) Evidence for landscape-level, pollen-mediated
gene flow from genetically modified creeping bentgrass with CP4 EPSPS as a marker. Proc
Natl Acad Sci USA 101:1453314538
Weimer PJ, Springer TL (2007) Fermentability of eastern gamagrass, big bluestem and sand
bluestem grown across a wide variety of environments. Biores Technol 98:16151621
Whitham TG, Gehring CA, Lamit LJ et al (2012) Community specificity: life and afterlife effects
of genes. Trends Plant Sci 17:271281
Wilsey BJ (2010) Productivity and subordinate species response to dominant grass species and
seed source during restoration. Restor Ecol 18:628637
Wu Y, Taliaferro C (2009) Switchgrass cultivar, cv Cimarron. US Patent Number: US
2009/0300977 A1
Wullschleger SD, Davis EB, Borsuk ME et al (2010) Biomass production in switchgrass across
the United States: database description and determinants of yield. Agron J 102:11581168
Yamada T, Forster JW, Humphreys MW, Takamizo T (2005) Genetics and molecular breeding in
Lolium/Festuca grass species complex. Grassl Sci 51:89106
Zalapa JE, Price DL, Kaeppler SM et al (2011) Hierarchical classification of switchgrass gen-
otypes using SSR and chloroplast sequences: ecotypes, ploidies, gene pools, and cultivars.
Theor Appl Genet 122:805817
Zapiola M, Mallory-Smith C, Thompson J, et al. (2007) Gene escape from glyphosate-resistant
creeping bentgrass fields: past, present, and future. In: Proceedings of the western society of
weed science, p 18
Zapiola ML, Campbell CK, Butler MD, Mallory-Smith CA (2008) Escape and establishment of
transgenic glyphosate-resistant creeping bentgrass Agrostis stolonifera in Oregon, USA: a
4-year study. J Appl Ecol 45:486494
Zhang Y, Zalapa JE, Jakubowski AR et al (2011a) Natural hybrids and gene flow between upland
and lowland switchgrass. Crop Sci 51:26262641
Zhang Y, Zalapa JE, Jakubowski AR et al (2011b) Post-glacial evolution of Panicum virga-
tum: centers of diversity and gene pools revealed by SSR markers and cpDNA sequences.
Genetica 139:933948
Zhang K, Johnson L, Nelson R et al (2012) Chemical and elemental composition of big bluestem
as affected by ecotype and planting location along the precipitation gradient of the Great
Plains. Industr Crops Prod 40:210218
Chapter 6
Mining Genetic Diversity of Sorghum
as a Bioenergy Feedstock
Cynthia M. B. Damasceno, Robert E. Schaffert and Ismail Dweikat
Abstract Sorghum is a drought-tolerant rainfed crop that requires about 30% less
nitrogen fertilizer than corn to produce equal amount of ethanol per acre under non-
irrigated conditions. Excellent genetic and genomic resources exist for improvement
of sorghum as a bioenergy source. We expect a huge impact on biomass yield, qual-
ity, and conversion efficiency with appropriate plant breeding and biotechnology
tools in order to develop energy sorghum germplasm that allows highly efficient pro-
duction of biofuel. The outlined improvement should produce benefits that include:
(1) genetic improvement of a biomass crop with significantly reduced overall cost of
biomass-to-ethanol conversion; (2) selection of a reliable bioenergy feedstock that
is drought tolerant, inexpensive to grow, environmentally friendly and cultivated in
nearly all temperate and tropical climate regions; (3) expansion of the production
area for bioenergy crops by developing cold tolerance germplasm and hybrids and
by offering both annual and perennial sweet sorghum types; and (4) reduction in cell
wall lignin for improved efficiency in production of biofuels.
C. M. B. Damasceno R. E. Schaffert
Embrapa Maize and Sorghum, Caixa Postal 151 Rodovia MG-424, Km 45,
34701-970 Sete Lagoas, MG, Brazil
e-mail: cynthia.damasceno@embrapa.br
R. E. Schaffert
e-mail: robert.schaffert@embrapa.br
I. Dweikat(*)
Department of Agronomy and Horticulture, University of Nebraska, 202 Keim Hall,
Lincoln, NE 68583-0915, USA
e-mail: idweikat@unlnotes.unl.edu

82 C. M. B. Damasceno et al.
6.1Introduction
Sorghum, [Sorghum bicolor (L.) Moench], is an excellent choice for a feedstock for
bioenergy and bioproduct production as it is a very photosynthesis efficient species
comparable to or more efficient than other C4 species such as sugarcane and ele-
phant grass. Sweet sorghum can be used for first generation technology (G1) which
involves the direct fermentation of the sugar produced in the stem and extracted in the
juice, similar to sugarcane. Highly productive energy sorghum or biomass sorghum
can be used as an efficient feedstock for second generation technology (G2) involving
the hydrolysis of the cellulose and hemicellulose to simple sugars and fermentation to
produce bioproducts and/or biofuels. Highly productive energy sorghum or biomass
sorghum can also be used as a feedstock for third, fourth generation and beyond tech-
nologies for generating bioenergy and bioproducts. Also, a strong demand for energy
sorghum is evolving in Brazil to burn the biomass for co-generation of electrical
energy or the generation of steam in certain industrial processes.
The top ethanol producers in the world are the United States and Brazil. A
research and development project to develop feedstock for G1 bioenergy was initi-
ated by the Brazilian Enterprise for Agricultural Research (Embrapa) in Brazil in
1975 as a response to the demand of the Pr-lcool Program of Brazil, following
the 1973 OPEC oil embargo. In response to the new energy demands Brazilian grow-
ing economy will require, the Brazilian government launched in 2006 the National
Agroenergy Plan in order to stimulate research and production in the bioenergy sec-
tor. In the United States (US), the Energy Independence and Security Act of 2007 was
established amending the renewable fuels standard (RFS)-Energy Policy Act of 2005,
in order to reduce its dependence on foreign fossil energy supplies, reduce green-
house gas emissions, and provide meaningful economic opportunity. By year 2050,
the energy demands in the US alone are expected to be about 27Terawatts per year
(Lewis and Nocera 2006). Biomass feedstocks are expected to contribute to as much as
40% of total renewable energy with expected biofuel production of 136 billion liters
in the US by 2022 (EIA 2011). In response to these demands, R&D projects have been
launched in recent years to develop adequate technology for the production of biofuels.
The advantage of plant-based biomass material lies in its photosynthetic ability.
Nature has designed a sophisticated solar conversion system that self-assembles from
water, nutrients in the soil and carbon dioxide (CO2) in the air with energy input
from the sun. Enhancing the production and conversion of plant biomass to utilizable
sources of energy is critical to decrease dependency on fossil fuels, increase energy
security, ameliorate negative environmental impacts associated with petroleum com-
bustion (e.g., reducing greenhouse gas emissions and toxic pollutants) and will add
significant value to domestic agricultural products. Converting plant biomass to etha-
nol represents a renewable form of energy that is immediately utilizable in existing
combustion engines and transportation systems (Sommervile 2007; Vermerris et al.
2007).
Sorghum by origin is a short day species in its native areas of evolution in
Tropical Africa or commonly called photosensitive (PS). In this case, flower induc-
tion is initiated when day length is less than 12h and 1520min (Rooney 2007).
6 Mining Genetic Diversity of Sorghum as a Bioenergy Feedstock 83
In the adaptation of sorghum to temperate environments, a mutation of the domi-

nant Ma1 maturity gene to ma1 was selected to develop photo insensitive sorghum
that will normally flower 6075days after germination. The sweet sorghum cul-
tivars currently being developed in Brazil and the US are photo insensitive (PIS)
and normally flower 6080days after germination and reach peak sugar concen-
tration around physiological maturity of the grain at 120130days after germina-
tion or 4070days after flowering. The months of late September to late March in
Brazil and the summer months in the US have days longer than 12h and 20min
providing a long growing season, up to 89months for energy sorghum. These
months coincide with the rainy period in the regions of the South East and Central
West of Brazil where the large distilleries are located and where the demand for
sweet sorghum and energy sorghum is evolving. The utilization of both PS and PIS
sorghum cultivars for bioenergy or bioproducts offers an array of opportunities for
providing feedstock for a range of industrial processes and activities.
Critical factors in utilizing biomass as an alternative energy source will be the
ability of the plant to achieve high biomass yields, grow in diverse climates under
various environmental conditions, and be economically converted into a bio-based
product. Synergistic improvements of biomass feedstocks, transportation and han-
dling logistics, and the efficiency of industrial conversion into bioenergy are essen-
tial for sustainable production of bioenergy. The major requirements for sustainable
biomass-based bioenergy production are high biomass yield, high energy content,
lower agronomic input requirements and high suitability for the end-user defined
processes, such as bio- or thermo-chemical conversion of biomass (Henry 2010).
Plant breeding is a cost-effective way to achieve an increased and stable yield.
Plant breeding allows for continuous increase and release of ever more productive
cultivars. In industrial terms, this increase will translate to, for example, higher
sugar content and juiciness that will lower the cost of making ethanol. The devel-
opment of multipurpose varieties will allow farmers to have additional markets
for their product (not just ethanol, it can be a food or feed and fuel crop at the
same time). The green revolution for major cereals would not have been made
possible without the release of outstanding varieties. A new green revolution will
require also new outstanding energy crop cultivars. Important features of potential
bioenergy feedstocks are; (1) the importance of deploying a non-food crop with
(2) capability for high growth potential on both fertile and marginal lands with
efficient water and nutrient utilization, (3) growth in a range of temperate to tropi-
cal climates, (4) amenability to production practices already in place for most
growers, (5) male sterility for production of hybrid cultivars or non-flowering
growth habit to maximize vegetative growth and minimize transgenic pollen
escape, (6) genetic tractability and availability of genomic resources for detailed
study, and (7) ability to supply carbon in a form that can be utilized within existing
bioenergy production schemes.
Sweet sorghum fits in very well as an additional feedstock to sugarcane in
the large distilleries in Brazil increasing the industrial processing period up to
6090days. Industrial processing of sugarcane begins in the month of April
or May until mid-December. The limiting factors for sugarcane are low sugar
content in April and high rainfall in November and December through January
and February complicating harvest and transport of the crop. Sugar content also
declines during the rainy season. Sweet sorghum can be planted at the beginning
or the rainy season in November and December for harvest and processing in
late February, March and April before the beginning of the sugarcane harvest in
April and May in areas of Sugarcane renovation. Sugarcane renovation is recom-
mended every five years in Brazil, thus generating a potential area of 20% of the
sugarcane acreage or 1.9 million hectares for sweet sorghum production between
the months of OctoberMay. Sustainable levels of ethanol production of 2,500
3,000lha1 in these conditions have been attained in four months in pilot trials
in Central Brazil (Embrapa Documentos 138). There are also other scenarios or
niches for sweet sorghum, such as in the implementation of new distilleries, areas
where sugarcane is restricted for environmental concerns, and in areas where rain-
fall is not sufficient for sugarcane production.
6.2Sorghum: Excellent Genetic and Genomic Resources

for Systems-Based Crop Improvement
Sorghum [Sorghum bicolor (L.) Moench] is the fifth most important cereal crop
world-wide (http://apps.fao.org/default.jsp) as well as an important source of food,
feed, fiber, and biofuel (Doggett 1988). Sorghum, like maize and sugarcane, car-
ries out C4 photosynthesis, a specialization that makes these grasses well adapted
to environments subject to high temperature and water limitation (Edwards et al.
2004). Different varieties of sorghum have been bred for different commodities.
Sweet sorghum has traditionally been bred for high sugar content in the stem for
producing molasses. These lines typically have high biomass and low seed yield.
On the other hand, grain sorghum has been bred for high seed yield, has reduced
plant biomass and low accumulation of sucrose in stem tissue. It is possible to
cross these varieties to generate populations that segregate for the various traits.
The gene for male sterility, Ms3/ms3 can be used for developing random mating
sweet sorghum populations for use in selection and improvement. The cross of
contrasting sweet sorghum lines to generate recombinant inbred lines (RILs) by
single seed descent is currently being used to map traits related to sucrose accu-
mulation and plant biomass at Embrapa.
Sorghum has cultivated varieties spread among 5 races and over 25 species
of wild relatives, providing tremendous genetic diversity for crop improvement
(Bhattacharya et al. 2011). Since its introduction to the Americas about 250years
ago, sorghum has been established as the third major cereal crop of both Brazil
and the U.S. During the last century, conversion of sorghum genotypes to adapt
to long day conditions has increased genetic diversity and greatly contributed
to improved grain crop quality and productivity (Marguerat and Bahler 2010).
However only about 1,000 of over 44,000 accessions of sorghum found at the
USDA National Plant Germplasm System (USDA-NPGS) flower under long day
conditions (photoperiod insensitive) suggesting that a vast untapped genetic diver-

sity is available for crop improvement for bioenergy traits.
Sorghum is also an important target of genome analysis among the C4 grasses
because the sorghum genome is relatively small (730Mbp) (Paterson et al. 2009),
the cultivated species is diploid (2n=20) and the sorghum germplasm is diverse
(Dje et al. 2000; Menz et al. 2004; Casa et al. 2005). As a consequence, numerous
sorghum genetic, physical, and comparative maps have been constructed (Tao etal.
1998; Boivin et al. 1999; Peng et al. 1999; Klein et al. 2000; Haussmann et al. 2002;
Menz et al. 2002; Bowers et al. 2003), a sorghum EST project (Pratt et al. 2005)
and associated microarray analyses of sorghum gene expression have been car-
ried out (Buchanan et al. 2005; Salzman et al. 2005). Also, a comprehensive analy-
sis of sorghum chromosome architecture has been done (Kim et al. 2005), and an
8x draft sequence of the sorghum genome (about twice the size of rice) has been
completed by the DOE Joint Genome Sequencing program (Paterson et al. 2009;
http://www.phytozome.net/sorghum). In addition, genetic maps have been assembled
at Texas A&M and University of Georgia (Menz etal. 2002; Bowers et al. 2003).
Several projects using the method genotyping-by-sequencing (GBS) are underway
at the Institute of Genetic Diversity at Cornell University, allowing large scale SNP
discovery for thousands of sorghum materials (lines and populations). Embrapa has
genotyped using GBS 275 F7 RILs developed from the cross of two sweet sorghum
varieties, Brandes and Wray and is in the process of phenotyping this population for
both productivity and quality. In addition to this, Embrapa has phenotyped a diverse
panel for biomass production and G2 quality which will be genotyped using GBS
as well, allowing for rapid and high resolution QTL mapping of bioenergy traits.
The whole genome sequence of sorghum BTx623 (Paterson et al. 2009) is a valu-
able reference for genome based discoveries in sorghum and existing and developing
genomics efforts complement these genetic resources and allow genomics-based
crop improvement for bioenergy traits.
6.3Sorghum Has Been Identified as a Preferred Biomass

Crop for Biofuels
Biomass can be used to generate electricity or to produce liquid transportation fuels.

Among the various types of renewable fuels (such as wind, solar, and geothermal),
biomass is unique because it is the only current renewable resource of liquid trans-
portation fuel. Sweet sorghum is a type of sorghum that has a high concentration of
soluble sugars in the juice, similar do sugarcane. Characteristics of high fermentable
sugars, efficient nutrient use, high water use efficiency (1/3 of sugarcane and 1/2 of
corn), short growing period with PIS and long growing season with PS sorghum cul-
tivars, and the ability to adapt well to diverse climate and soil conditions make sweet
sorghum a potential feedstock for ethanol production (Wu et al. 2008). While single-
cut yields may be lower, an increased growing season increases cumulative yields in
PS cultivars and the ratoon potential of PIS cultivars (Rooney et al. 2007).
Sorghum is among the most widely adaptable cereal grasses potentially useful for
biomass and fuel production (Hons et al. 1986). The adaptation of sorghum to sub-
humid and semiarid climates has extended the geographic scope of sorghum pro-
duction far beyond that of other warm-cereal grains. Sorghum is a versatile crop,
one grown in warm and cool climates. This annual C4, self-pollinating, highly water
efficient plant of African (tropical) origin is also well adapted to sub-tropical and
temperate regions. Sweet sorghum is a warm-season crop that matures earlier under
high temperatures and short days. It is not only known as a high energy crop for
its high photosynthetic rate, but it is also the camel among crops for its drought
and aluminum toxicity tolerance. Sweet sorghum is adapted to widely differing cli-
matic and soil conditions. As a consequence, sorghum plays a vital role in the global
food economy and is the fifth most important cereal cropfollowing wheat, rice,
corn, and barleywith more than 43 million hectares planted to sorghum worldwide
each year (FAO 2011). Sweet sorghum varieties can grow up to 5m tall and produce
50120tons of biomass (fresh weight) per hectare. Biomass, sugar extraction, and
ethanol production from traditional and new varieties of sweet sorghum ranged from
45 to 52Mgha1, 72 to 109kgMg1 and 1,800 to 2,600Lha1, respectively in
Brazil (Schaffert et al. 1986). Sweet sorghum requires less than 50% total nitrogen
to produce similar ethanol yields as corn (Anderson et al. 1995) and is capable of
removing 62% of total nitrogen with no difference in DM yield (Bean et al. 2008).
Reports have shown that sweet sorghum yielding 1116Mgha1 will remove
nitrogen, phosphorus and potassium at the rate of 112, 45, and 202kgha1, respec-
tively (Undersander et al. 1990). Under favorable conditions, sweet sorghum is
capable of producing up to 13.2 metric tons per hectare of total sugars, which is
equivalent to 7,682l of ethanol per hectare (Murray et al. 2009). Sorghum stalks
can reach a height of 5m with diameters ranging from 1 to 5cm. Sorghums small
leaf surface and very developed root structure (twice that of corn) are likely respon-
sible for the plants exceptional drought tolerance. Sorghum has also been reported
to tolerate fungal disease, viruses, herbicides, heat, insects, weeds, and poor quality
(alkali or acid) and water logged soils. Most of the sugars (sucrose, fructose, and
glucose) are uniformly distributed in the stalk, while about 2% are in the leaves and
inflorescences (Vietor and Miller 1990). Even in dry climates, sorghum can yield
high levels of fermentable sugar, grain, and lignocellulose (Gnansounou et al. 2005).
6.4Sweet Sorghum is Ideal for Both Sugar- and Cellulosic-

Based Ethanol
Sorghums can be classified into four main groups depending on their production
characteristics: grain sorghum, forage sorghum, high-tonnage sorghum or energy
sorghum, and sweet sorghum. Although sweet sorghum, as a crop, meets the needs
of US Midwest growers, grower acceptability of dedicated energy crops will be
greatest if the crops can be planted and harvested with the machinery used for cur-
rent crops, if they can be easily eradicated should landowners want to change land
use, and if they can provide harvestable material in a short period of time. Sweet
sorghum fits all of these requirements. Sorghums high water and N use efficiency
will further enhance farmer acceptability.
Sweet sorghum is characterized by high fermentable sugar content. At the soft
dough stage, the sugar is composed mainly of sucrose (6074%) (Table6.1), fruc-
tose and glucose that can be easily fermented to produce ethanol. Sweet sorghum
total fresh weight yields vary considerably from 12 to 60Mtha1 with arrange
from depending on the cultivars/hybrids used, the location, inputs, and production
practices (Dweikat et al. 2012; Jackson et al. 1980; Reddy et al. 2007; Propheter
et al. 2010; Zhao et al. 2009). Sweet sorghum sugar yields range between 1.6 and
13.2Mgha1, with significant variation observed across different years and regions
(Jackson et al. 1980; Reddy et al. 2007; Propheter et al. 2010; Zhao et al. 2009).
Cellulosic materials are generally thought to be the most likely feedstock for large-
scale ethanol production from biomass in the long-term, due to their potentially larger
supply and lower price compared to other carbohydrate sources (Perlack et al. 2005).
However, low cost, plentiful supply, and ease of conversion have made readily ferment-
able carbohydrates (FC) the preferred feedstocks for bioethanol production. Starch-rich
materials, such as grains, have the advantage of established feedstock and processing
infrastructure in the US, and a more homogenous and reactive form of carbohydrate
than that found in cellulosic materials. An advantage of both starch and sugar-rich
materials over cellulosic materials is that they can be processed to sugar streams of suf-
ficient purity to accommodate production of high-value products such as food, phar-
maceuticals, and fiber-grade polymers. Plant materials high in soluble sugars yield the
most readily converted form of carbohydrate, requiring lower inputs of chemicals and
energy for processing, and the technology for the extraction of sugars is fully mature
and highly efficient, reducing processing costs. Sugar is the preferred carbohydrate
feedstock for many high-value products and is also used to produce around half of the
worlds ethanol, the largest bio-commodity (Murray 2005).
Sweet sorghum is of particular interest because of the large volume of read-
ily fermentable juice that can be extracted. Hunter and Anderson (1997) indicated
Table6.1Properties of Character Variety Variety

sweet sorghum at the soft
Cultivar Simon M 81E
dough stage, including stalk,
leaves, panicle and grain, Moisture content 6570 6570
for two varieties. Except Sugar 3542 3540
for moisture, % is in terms Fiber 1315 1315
of dry mass. Sampling was Cellulose 3337 3237
done field planted at 70,000 Hemicellulose 2024 2024
plants/acre Protein 47 58
Aconitic acid 24 24
Starch (Juice) 0.54 0.54
Oil and wax 34 34
Ash 24 24
Total 100% of dry matter 100% of dry matter
that the sugar produced in sweet sorghum has a potential ethanol yield up to
8,000Lha1, or about twice the ethanol yield potential of maize grain, In addition
to producing large amounts of sugar-rich biomass, hybrids can be developed from
crosses between grain-type seed parents and sweet-type pollen parents (Hunter
and Anderson 1997). The product of these crosses typically increase biomass
yields and sugar content when compared to the original grain-type seed parents,
but are inferior in sugar quality compared to the pollinator parent. Such hybrids
can co-produce grain at levels approaching the yields of the grain-type seed par-
ent (Miller and McBee 1993). Sweet sorghum has been found to be competitive
with corn for theoretical ethanol yield with less energy invested (Smith et al. 1987;
Smith and Buxton 1993; Hunter and Anderson 1997). Ethanol production from
sugar does not require energy to depolymerize carbohydrates such as is required
for grain starch or cellulosic ethanol. Smith et al. (1987) reported total sugar yield
ranging from 4 to 10.7Mgha1 for the continental USA and up to 12Mgha1
for Hawaii while Smith and Buxton (1993) reported sugar yields at 6Mgha1 in
Iowa and Colorado. Vermerris et al. (2007) reported total sugar concentrations of
the juice ranging from 9 to 15%. Ricaud et al. (1979) found sugar concentration
in juice at the soft dough stage to range from 12.8 to 16.6%. Theoretical ethanol
yield estimates for SS have ranged from 3,850 to 4,410Lha1 (Lueschen et al.
1991; Hunter 1994) although the crop has been estimated to have the potential of
8,000Lha1 of ethanol (Hunter and Anderson 1997), equivalent to ethanol pro-
duced from approximately 20Mgha1 of corn.
The conversion efficiency of lignocellulosic biomass to liquid fuels like ethanol
is strongly influenced by cell wall composition. Lignin has been shown to hamper
saccharification by physically shielding the cellulose from degradation, making it
difficult to convert into ethanol and increasing the energy requirement for process-
ing. Sorghum stover, containing lignin, hemi-cellulose and cellulose (Table6.1),
can serve as an excellent feedstock for ethanol production. A set of mutation
stocks, developed by the USDA Plant Stress and Germplasm Development Unit in
Lubbock, Texas, USA (Xin et al. 2008), is sufficiently extensive to allow identifica-
tion of mutations in virtually every sorghum gene. The best known such mutations
are the brown midrib (BMR) mutants, which were first discovered in maize in 1926.
Early studies revealed the trait resulted in lower fiber and lignin within the plant
and could increase the conversion efficiency of sorghum biomass for lignocellulosic
bioenergy. In sorghum, more than 19 bmr mutants were discovered by Porter et al.
(1978). The bmr mutants are characterized by the reddish-brown coloration of the
vascular tissue of the leaf blade, leaf sheath and stem, which is associated with alter-
ation of secondary cell wall composition, especially lignin.
The bmr mutant sorghum, pearl millet (Pennisetum glaucum) and maize lines
have significantly lower levels of lignin content (51% less in their stems and
25% less in their leaves). Purdue University research showed 50% higher yield
of fermentable sugar from the stover of certain sorghum bmr lines after enzymatic
hydrolysis (Vermerris 2011). Therefore, the use of bmr cultivars would reduce the
cost of biomass-based ethanol production. In addition, The bmr crop residues have
higher rumen digestibility and palatability, making them good for fodder as well.
6.5Sorghum Production
Sorghum is relatively inexpensive to grow to high yields, and can be used to

produce high value-added products like ethanol and distillers dried grains
(Chiaramonti et al. 2004). Due to its high productivity and rapid growth cycle
(120150days), sweet sorghum has an impressive capacity to absorb a large
amount of CO2 from the atmosphere during its growing cycle. When compared
to the input requirements of other crops, sorghum requires half of those needed
by sugar beets, and one-third of the requirements of sugar cane or corn (Soltani
and Almodares 1994; Renewable Energy World 2000). The cost of production per
unit area e.g. one hectare is about $250 less than corn due to seed cost and lower
requirement of fertilizers. Bennett and Anex (2009) compared the production,
milling of sweet sorghum to corn and concluded that when combustion credits are
$68GJ1, sweet sorghum costs are in the range of $91149Mg1 compared to
$171258Mg1 for corn. Sweet sorghum juice is ideally suited for ethanol pro-
duction given its higher content of total reducing sugars compared to the content
of other sources, including sugarcane juice. Further, following the extraction of
juice, sorghum bagasse can be burned to generate steam for ethanol distillation
and co-generation of electricity. Remaining bagasse can be used as fodder for ani-
mals or for additional ethanol production through lignocellulose conversion.
Also important is the amount of energy used to produce ethanol. Historically, for
each unit of energy it took to plant and harvest a crop and process it into ethanol,
the fuel returned 0.92 units of energy. Ethanol had a negative energy balance of
one unit in for 0.92 out (1:0.92). However, steady improvements have been made
in corn yield and harvesting and in ethanol processing efficiency. The latest studies
show corn ethanol boasts a positive energy balance of 1:1.25; a 25% net increase
in net energy (Farrell et al. 2006). Today, corn ethanol is made by converting the
starch in corn to sugars and then into alcohol by a fermentation process. Sugar beets
(Beta vulgaris L.) are a better ethanol source, producing nearly two units of energy
for every unit used in production. Sugarcane, though, is by far the most efficient of
the current feedstocks, yielding more than three units as much energy as is needed
to produce the ethanol derived from it (Hopkinson and Day 1980). Sweet sorghums
positive energy balance, with a ratio of 1:8, is comparable to that of sugarcane
(Worley et al. 1992). Given their positive energy balances and higher yields, it makes
more sense to produce ethanol from sugar crops than from starchy grains.
6.6Sweet Sorghum Does Not Compromise Food Security
Corn is currently the feedstock of choice for U.S. ethanol producers. Increasing
ethanol production led to higher domestic corn utilization, as it is also widely
used in the food and livestock sectors. This, coupled with other factors such as
the value of the dollar and investment markets, has contributed to corn prices ris-
ing to some of the highest levels in U.S. history. Farmers responded to high corn
prices by shifting planted acres to corn, which has caused ripple effects across
other crops, contributing to higher price levels of competing crops. As a result,
public and political interest has escalated for the production of ethanol from
sources other than corn. Economic research has explored various alternative etha-
nol production technologies. Progress has been made with respect to biochemical
and thermo-chemical technologies for cellulosic ethanol, yet the ability to reach
commercial viability continues to elude the industry. Herbst (2003), Shapouri et al.
(2006), Ribera et al. (2007), Salassi (2007), and Outlaw et al. (2007) have exam-
ined the economic feasibility of ethanol production from grain sorghum and corn,
sugarcane juice and molasses, respectively. Studies by Epplin (1996), Graham
et al. (2000), and Mapemba et al. (2007) have explored transportation, harvest,
and delivered feedstock cost components of biomass used for cellulosic ethanol.
Outlaw et al. (2007) conclude ethanol production from sugarcane juice, a pre-
dominant production method in Brazil, would be economically feasible in certain
regions of the United States. However, sugar policy has left little opportunity for
this method to gain traction in the United States.
6.7Sweet Sorghum is Ideal for Double Cropping
Because sorghum grown for biomass can be harvested before it is fully mature,
it is possible to grow it in a double crop sequence with a winter annual. Winter
annuals are planted in the fall, grow rapidly in the spring, and reach harvest any-
time during late spring to early summer. Sorghum, which is well adapted to ger-
mination under limited moisture, can be no-till planted into the stubble of a winter
annual crop. The primary advantages of a double crop sequence are to maximize
use of solar radiation, provide winter cover against wind and water erosion, and
increase biomass/ha. Because sorghum and winter annuals have differing cardinal
temperatures for growth, the double crop sequence can take advantage of a longer
growing season than either crop alone.
Sweet sorghum yield has not been found to be sensitive to plant density
(Lueschen et al. 1991; Ferraris and Charles-Edwards 1986), but sucrose con-
tent, sugar yield, and juice content have been decreased with high plant den-
sity (Broadhead and Freeman 1980; Martin and Kelleher 1984; Kuepper 1992).
Efficient nitrogen (N) use is important for net energy yield (energy yield rela-
tive to energy invested), and associated life cycle greenhouse gas (GHG) emis-
sions (Liska and Cassman 2008). Response to applied nutrients has varied with
location. Nitrogen application did not affect fermentable sugar yield (Smith and
Buxton 1993), total and stalk dry matter yield at harvest (Barbanti et al. 2006), or
fermentable carbohydrate and ethanol yield (Lueschen et al. 1991). Biomass yield
has been shown to increase in Louisiana by 140% by applying 100kgha1N,
but no further increase was observed with an additional 100kgha1N (Ricaud
and Arenneaux 1990). Total sugar yield was also increased by 150% in the
same study by applying 100kgha1N, with a 4% increase from an additional
100kgha1N. Sweet sorghum has been found to require roughly 36% of the
fertilizer N required for similar yield levels in corn (Geng et al. 1989). Total dis-
solved solid concentration in stalk juice was shown to decrease with increased N
rate (Wiendenfeld 1984). Consequently, some priority areas of research for sweet
sorghum as an ideal bioenergy crop are the annual nature of the crop in the tem-
perate regions, sensitivity to cool temperatures at the early growing stages, weed
controls for industrial scale plantation, sustainable production systems.
6.8Development of Perennial Sorghum
Perennial plants are highly efficient and responsive micro-managers of soil, nutri-
ents, and water. In contrast, annual crops require seedbed preparation, precisely
timed inputs and management, and good weather during narrow time windows.
With shorter growing seasons and less extensive root systems, annual crops pro-
vide less protection against soil erosion, manage water and nutrients less effec-
tively, store less carbon below ground, and are less resilient to pests and abiotic
stresses than are perennial plant communities (Glover 2005). Perennials generally
yield more above ground biomass than do annuals, and some of the carbon that
goes into the biomass might be reallocated to above ground section of the plant
production through breeding. Although those species currently being domesti-
cated as perennial grain crops have low seed yields, their total aboveground pro-
ductivity is often higher than that of annual crops with long breeding histories
(DeHaan etal. 2005). Piper and Kulakow (1994), for example, reported a mean
aboveground biomass for self-pollinated progeny of annual X perennial sorghum
hybrids that was 62% higher than that of their annual parent.
The perennial species Sorghum halepense is a tetraploid with 20 pairs of chro-
mosomes, 10 of them similar to those of S. bicolor and 10 similar to those of
the diploid perennial S. propinquum (Paterson et al. 1995). Sorgum halepense as
migrated throughout much of the United States and into Canada as a highly suc-
cessful, rhizomatous weed known as johnsongrass. Because the two species differ
in chromosome number, those early hybrids were produced using artificially chro-
mosome-doubled (tetraploid) S. bicolor plants as female parents. The tetraploid F1
hybrids were fertile, and F2 plants derived by self-pollination of the hybrids varied
widely in rhizome production. A majority of those plants were perennial; that is
after harvest, their rhizomes survived through the winter to produce new shoots the
following spring. Advance sorghum lines with rhizomes have been generated as a
result of crossing grain sorghum nuclearmale sterile lines to Johnsongrass mate-
rials (Dweikat 2005, Fig.6.1). The sorghum lines were backcrossed five genera-
tions, and grain types with rhizomes habit were selected. Our goal is to use these
sorghum lines for crosses with elite sweet sorghum cultivars. The F1 plants were
backcrossed to sweet sorghum to generate sweet sorghum advanced lines with
rhizomes. The advanced materials have been planted in the field and have been
allowed to over-winter to select for perennial biotypes.
S. bicolor F1 S.halapense
Fig. 6.1a Chromosome analysis of root-tip cells collected from Johnsongrass (right), culti-
vated sorghum (left) and the putative F1 plant (middle) confirmed that Johnsongrass contained
40 chromosomes while both the cultivated sorghum and the F1 contained 20 chromosomes.
b Phenotypic characterization of the seed (top panel), panicles (middle panels) and roots
(bottom) of sorghum (left), Johnsongrass (right), and the F1 hybrid (middle)
6.9Breeding for Cold Tolerance as a Mean to Extend

Sorghum Production Regions
Sorghum originated from tropical regions and is a cold and frost-sensitive plant.
Low-temperature stresses, including chilling and frost, greatly affect the germina-
tion and growth of the plant, and limit the geographical distribution of the crop.
This poses a major problem in temperate environments when sorghum is planted
early. Cool air and soil temperatures result in poor seedling establishment of sor-
ghum because of slow emergence rate, reduced emergence percentage, and reduced
growth rate after emergence (Pinthus and Rosenblum 1961; Singh 1985). Early
planting of sorghum hybrids in temperate climates confers numerous advantages,
but requires genetic improvement, both in lines and in parental effects for germi-
nation and seedling tolerance to low soil temperatures. Analysis of the phenotypic
data showed a high degree of heritability for all traits measured (Gunaratna 2002),
suggesting that gains from selection for seedling cold tolerance should be signif-
icant. The development of sorghum hybrids with increased cold tolerance would
help to expand and stabilize sorghum production under these conditions. Sorghum
hybrids with early-season cold tolerance have several advantages. First, early can-
opy development reduces weed competition and soil moisture evaporation. Second,
sorghum primarily is grown in hot and dry environments, and early planting allows
the crop to better utilize spring rains and reach flowering prior to drought condi-
tions that are more prevalent during the mid-summer period (Pendleton et al.
1965). Finally, improved cold tolerance would be important in no-tillage produc-
tion systems. No-tillage systems are becoming increasingly popular due to their
numerous advantages, but they also reduce soil temperature by 15C at typical
planting dates (Carter and Barnett 1987; Graven and Carter 1991).
6.10Development of Herbicides Tolerance
Sorghum has the ability to tolerate short-term drought and a late summer sorghum
crop may follow an early-season corn crop. Achieving good control of grassy
weeds has been identified by producers as a significant management challenge that
must be addressed in order for the crop to be economically sustainable. New her-
bicides are not developed specifically for grain sorghum. If a herbicide developed
for corn or wheat does not cause severe phytotoxicity in grain sorghum, then grain
sorghum may be added to the label. In fact, because of the high cost of develop-
ing, testing, and registering herbicides, there are very few new herbicides being
developed now even for the major crops. Weed control in sorghum is essential if
high yields and efficient harvest are to be achieved; however, good weed control
in sorghum is often difficult to achieve. Sorghum is a small seeded grass and is
relatively slow growing in the first few weeks after emergence. The slow seedling
growth combined with the limited number of herbicides and the low rates which
must be used creates a problem in sorghum weed control.
The most troublesome weeds for grain sorghum include morning glory, pig-
weed, broadleaf signal grass, barnyard grass, prickly sida, crabgrass and sick-
lepod (referencesthis is not uniform in all sorghum growing areas). There are
fewer control options for weed control in grain sorghum than in corn, cotton and
soybeans. Preemergence and POST ALS-inhibiting herbicides are used effec-
tively to control weeds in corn (Zea mays L.) and other crops. Unfortunately,
sorghum is susceptible to grass control ALS-inhibiting herbicides such as nicosul-
furon and rimsulfuron, which makes it impossible to use these herbicides in sor-
ghum. Recently, researchers at both Kansas State University and the University
of Nebraska have developed different types of sorghum that is resistant to several
ALS-inhibiting herbicides by transferring resistance genes from a wild sorghum
relative (Tuinstra and Al-Khatib 2007; Tuinstra et al. 2009; Hennigh et al. 2010;
Gelli et al. unpublished).
Shattercane (Sorghum bicolor) is a monocot weed in the Poaceae fam-
ily. It occurs as a weed in wherever cultivated grain sorghums and their wild
relatives grow in the same region. All races of the subspecies bicolor and all
wild kinds of S. bicolor can hybridize into the weedy shattercanes. Four differ-
ent shattercanes resistant to ALS-inhibiting herbicides were collected between
1992 and 1996 from plants in 16 fields located in southeastern and south cen-
tral Nebraska. The plants were randomly selected from those fields which had
been treated for three consecutive years with different classes of ALS inhibiting
herbicides. Greenhouse experiments will be conducted to evaluate the response
of the four different putative resistant shattercane biotypes to ALS-inhibiting
herbicides. The resistant and the susceptible biotypes will be tested against the
four classes of ALS-inhibiting herbicides SU, PTB, TP, and SCT (Gelli et al.
unpublished).
6.11Breeding Strategies for Sweet Sorghum
Historically, sweet sorghum cultivars have been photo insensitive varieties.

Sweet sorghum varieties were initially introduced into the USA in the 1850s
and developed in the United States in the 1880s and 1890s to develop sweet
syrup, especially when crystal sugar was unavailable. Research investment
continued with the development of improved varieties through the 1970s and
1980s. The sweet sorghum varieties developed for high quality syrup produc-
tion were selected for high Brix and high total sugars in the juice, but reduced
levels of sucrose to avoid crystallization of the syrup. Sweet sorghum syrup that
crystallized was considered to have reduced quality. High sucrose sweet sor-
ghum varieties such as Rio, Roma, Ramada, Keller and Wray were developed
and released in the 1970s and 1980s with the objective of producing crystal
sugar in the sugar mills of Southern Texas and Mexico. The high level of starch
(up to 5,000ppm) in the juice in high sucrose varieties interferes in the sucrose
crystallization process.
Table6.2Sweet sorghum minimum yield and quality goals

Trait Target 1975 Target 2013
Minimum biomass yield 40tha1 5060tha1
(Genetic potentialproduction system) (10tha1 month) (1215tha1 month)
Minimum brixa 1517 1619
Peak brixa 21 23
Minimum total sugar extracteda (kgt1biomass) 80 100120
Sucrose (POL) 1018
Purity (% Sucrose) 7090%
Juice extraction efficiency 6065% 9095%
Minimum total sugar (ART) content in juice 12.5% 14%
Minimum alcohol yield 40lt1biomass 6070lt1
Minimum alcohol yield 2,500lha1 3,500lha1
Fermentation efficiency (%) 90 95
Distillation efficiency (%) 90 95
Industrial efficiency (%) 81 90
Period of industrial utilization (PIU) 30days 30days
Panicle size Small
Tillering Non-tillering
Cultivar type Variety Variety/hybrid
a Based on standard hydraulic press extraction of 500g sample with 245kgcm2 for 60s
Embrapa conducted a sweet sorghum breeding program for ethanol production

between 1975 and 1987 and reinitiated this R&D activity in 2008. The original and
updated breeding objectives of Embrapa for developing sweet sorghum cultivars
for ethanol production in Brazil are summarized in Table6.2. The biomass yield is
a function of the genetic potential of the cultivar used and the production system
and level of production technology utilized. Experimental yields of total biomass of
sweet sorghum varieties developed by Embrapa have surpassed 100Mgha1. The
Brix is a measure of soluble solids in solution including sugars and soluble salts.
In sweet sorghum with high levels of Brix (1621), the total sugar content is nor-
mally approximately 11.5 units less than Brix and at lower levels of Brix (812),
the sugar content is several units less than Brix, frequently 46% sugar. The sugar
extracted is a function of biomass yield, fiber content (related to juice extraction)
and juice quality. The presence of sucrose in the total sugar content is not essen-
tial for G1 ethanol production. However, higher levels of sucrose appear to be cor-
related (Non-published data Embrapa) with both higher total sugars and longer
periods of industrial utilization (PIU), the minimum period of adequate quality for
efficient and sustainable industrialization (PIU is discussed in more detail below).
Consequently, a breeding strategy for G1 sweet sorghum should include selection
for high sucrose (POL) and high purity (sucrose as a percent of total sugars). The
juice extraction efficiency is based on the fiber content of the sorghum stalks and the
efficiency of the equipment used in processing the feedstock and may vary depend-
ing upon the size and efficiency of the distillery. The minimum level of total sugars
(12.5%) in the juice is governed by the efficiency of the yeast and tolerance of the
yeast to the level of ethanol and may be increased as progress is made in selecting
more efficient yeast strains more tolerant to higher ethanol concentration. Juice with
total sugar content higher than 12.514.0% sugar content can be diluted with water
for efficient fermentation. The alcohol yield per ton of stalks is a function of extrac-
tion efficiency, sugar content in the juice, and industrial efficiency, and alcohol yield
per hectare is a function of biomass yield, extraction efficiency, sugar content in the
juice and industrial efficiency.
The above discussion is based on a standardized process of determining Brix,
sugar content of the juice and total sugar extracted. We have used the hydrau-
lic press (http://www.pontalmaq.com.br/links/sub_link/prensa_hidraulica.html)
developed and utilized by the sugarcane industry to characterize experimental
germplasm and commercial cultivars. The Brix measurement from a field sam-
ple can vary significantly depending upon how the juice was extracted. The Brix
reading of a few drops of juice (easy juice) on a digital refractometer in the field
will be several units higher than the Brix reading of the juice extracted with
a hydraulic press (minimum of 810 stalks). Consequently, one must beware
and knowledgeable of how the juice was extracted for proper interpretation.
Economic ethanol yield projections based on easy juice samples will be very
misleading by projecting exaggerated ethanol yield productions. Also, one must
keep in mind that Brix is not sugar and cannot be used directly to substitute
total sugar levels in estimating ethanol production as has been reported in sev-
eral publications [Degrees Brixsymbol Bxis the sugar content of an aque-
ous solution (http://www.engineeringtoolbox.com/degrees-brix-d_1828.html)].
6.12Industrial Management
The period of industrial utilization (PIU) is terminology borrowed from the sugarcane
industry to describe the time interval that a cultivar of sweet sorghum has the mini-
mum levels of sugar extraction and minimum level of sugar in the juice for economic
production of ethanol. Experience at Embrapa, based on data utilizing the hydraulic
press, indicates that a minimum total sugar of 12.5% in the juice and extracted sugar
of 80kgt1 biomass over a period of 30days are the minimum parameters for deter-
mining the PIU. Normally total sugar of 12.5% and extracted sugar of 80kgt1 bio-
mass occur simultaneously. The PIU of a cultivar can vary at different planting times
and at different locations and should be determined for each eco-region where the
cultivar will be utilized. This information is necessary for industrial planning in order
to provide the necessary amount of biomass daily for the programmed harvest period.
The logic of using a minimum period of 30days is so that the industrial manager can
program each cultivar and planting period for a 15day harvest window, thus allowing
a 15day cushion for providing the necessary quantity of daily feedstock for milling.
A maturity curve, the interaction of juice extraction, fiber, and sugar content of the
juice, is developed by weekly sampling of a cultivar beginning a few days (15) after
flowering for a period of 810weeks. An example of a sweet sorghum maturity curve
is presented in Fig.6.2.
Sweet sorghum maturity curve for determining

Period of Industrial Utilization
20 (80)
BR505 - Wray
Total Sugar Extraction (kg.t -1 biomass)

Brix, total sugars %, fiber %,
Brix (% juice)
Sugar extractio
n (%
(juice extraction %)
juice) biom
ars (% ass
)
15 (70) a l sug 90
Tot b iom a ss )
Fibe r (%
Juic
80
e ex
trac
tion
10 (60) (% b 70
ioma
ss)
60
5 (50)
85 90 95 100 105 110 115 120 125 130 135 140 145 150
Days after planting
The interaction of Brix and total sugars in the extracted juice, fiber,
juice extraction and sugar extraction of sweet sorghum stalks
during maturity for the variety Wray utilizing a hydraulic press.
(Embrapa Maize and Sorghum, Sete Lagoas, MG, Brazil)
Fig.6.2Sweet sorghum maturity curve for determining period of industrial utilization
6.13Sweet Sorghum Idiotype
The ideal sweet sorghum idiotype is a non-tillering cultivar with a small panicle
(Fig.6.3) with the parameters in Table6.2. The number of stems per hectare influ-
ences biomass yield, biomass quality, and lodging. The only way to control stem
population is by controlling the seeding rate of non-tillering cultivars. Small pani-
cles are desirable as this reduces lodging, provides less competition for photosyn-
thates and reduces harvesting and processing complications.
Fig.6.3Ideal sweet sorghum

idiotype
Table6.3Minimum thresholds recommended for sweet sorghum in Brazil

19752011 2012/2013 Ideala
Biomass production (t ha1) 40 50 6070
Total sugar in juice (%) 12.5 12.5 >14.5
Ethanol production (l t1) 60 60 70
Ethanol production (lha1) 2,500 3,000 4,2004,900
Period of industrial utilizationPIU (days) 30 30 30
aIdeal threshold of a non-photosensitive (PIS) sweet sorghum cultivar in Brazil (Embrapa maize
and sorghum)
6.14Varieties Versus Hybrids
Both hybrids and varieties of sweet sorghum cultivars have been evaluated in
experimental trials and pilot evaluations in Brazil. The varieties were developed
by Embrapa and the hybrids were developed and provided by several private seed
companies. For the most part, no differences were observed in biomass produc-
tivity, with both varieties and hybrids producing acceptable levels of biomass of
more than 40tha1. However, for the most part, the sugar quality of the hybrid
biomass has not reached the minimum thresholds (Table6.3) of sugar quantity, for
both sustainable processing and sustainable ethanol yields. The varieties, on the
other hand, produced both adequate biomass quantity and quality when adequate
production systems are used, reaching the minimum ethanol production of 2,500
3,000lha1. Research at Embrapa has indicated that the lower industrial quality
of hybrids can be contributed to the low sugar level of the female lines. No differ-
ences have been observed in the quantity of juice extracted between varieties and
hybrids, but the Brix and sugar levels in the juice of the hybrids is only slightly
superior to the average or mid-parent of the male and female parents. This has also
been reported by several authors (Reddy et al. 2011).
The presence and absence of juice in the stem of sorghum is simply inherited
(one recessive gene) and many juicy stem female lines are used in producing both
grain and forage sorghum hybrids. Sugar in the juice, on the other hand, is a more
complexly inherited trait with several genes being involved. There is a very lim-
ited number of short statured three dwarf sweet sorghum female lines available
to produce sweet sorghum hybrids. Another complicating factor to develop ideal
sweet sorghum lines is that all female lines developed since the 1950s have been
selected for high general combining capacity (GCC) for grain production, whereas
in sweet sorghum the ideal type is minimum grain production. The rational for this
is to reduce lodging, to simplify the mechanical harvest of sweet sorghum biomass
for producing ethanol using G1 technology, and to reduce possible competition
for sugars in the photosynthesis process. We have observed that in experimen-
tal hybrids, both an increased amount of stalk lodging occurs and lower levels of
sugar in the juice resulting in an inadequate and shorter PIU.
The most recently released cultivars (USA), cultivars released in the 1970s
and 1980s such as Topper, M81E, Wray, Theis, Rio, etc. were improved varieties.
A small emphasis was placed on developing cytoplasmic male-sterile female A

and B lines for the A1 cytoplasm by some public institutions in the USA, but
desirable results with hybrids have not been documented using these female lines.
The experience of Embrapa with sweet sorghum hybrids in Brazil is very similar
to the released and experimental hybrids of the private sector in that the sugar
quality has not met the minimum standards of the distilleries. In collaboration
with the distilleries, Embrapa has established the concept of minimum ethanol
yield per hectare. The private sector distilleries initially established a minimum
ethanol yield of 2.500lha1 in 2011 and have suggested in 2012 and 2013 that
an ideal minimum ethanol yield of 3,000lha1 provides a more adequate profit
scenario. The first variety released by Embrapa, BR506 has met these mini-
mum requirements in pilot operations at large distilleries. The varieties released
by Embrapa in 2012, BRS508, BRS509 and BRS511, also meet these minimum
requirements.
Sweet sorghum varieties are the preferred cultivars; however, hybrids will
be the cultivars of the future because of the possibility of mechanized seed
production to produce the large quantity of seed necessary to meet the pro-
jected demand. Also, the private sector prefers hybrids to protect their intel-
lectual property. The path to successful development of high yielding, high
quality sweet sorghum hybrids is to develop short statured three or two
dwarf sweet sorghum high quality male sterile lines suitable for mechanical
harvesting.
6.15Strategy for Developing Cytoplasmic Male Sterile

Sweet Sorghum Lines
The principal cytoplasmic male sterility utilized in developing male ster-

ile (A and B lines) and restorer (R) lines is the A1 cytoplasm. The fertil-
ity restorer lines (R-lines) depend upon at least one of three dominant genes,
Rf1/rf1, Rf2/rf2 and Rf5/rf5. The recovery of B-lines from BR crosses
that give 100% male sterility when backcrossed into the A1 sterile cyto-
plasm is much lower than from BB crosses. The great majority of the
sweet sorghum varieties developed over the past several decades are R-lines.
Embrapa has used the elite R-line, Wray, and Wray derivatives in crosses with
elite juicy stem B-lines to develop high sugar, high sucrose, non-tillering,
small panicle size B-lines for use in developing ideal sweet sorghum male sterile
A-lines for use in developing high quality sweet sorghum hybrids with the desir-
able characteristics described in Table6.2. The first set of high sugar female lines
is currently being evaluated at Embrapa for 100% male sterility in the A lines and
in hybrid combinations.
Over the past 2years, Embrapa has identified 12 sweet sorghum non-restorer
cultivars from its sweet sorghum germplasm collection. These lines have been
characterized and the best lines are being utilized in BB crosses with elite multi-
ple stress tolerant juicy stem B-lines to develop a broad based array of male sterile
lines for developing sweet sorghum hybrids.
6.16Breeding Strategies for Energy Sorghum
The basic objective for developing energy or biomass sorghums is to maxi-

mize the productivity. This is best achieved by using photosensitive sorghum
hybrids and it can be achieved in Brazil using two different models. The first
and most efficient model is using photo insensitive (PIS) female lines with the
recessive maturity gene ma1ma1 and PS restorer lines with dominant maturity
gene Ma1Ma1 as the male parent. These two types of sorghum will have syn-
chronized flowering when planted in March and April, facilitating commercial
seed production. Total biomass production of 60tha1 dry weight has been
observed for experimental PS hybrids in experimental biomass trials. Pilot
production of PS experimental hybrids has been over 50tha1 dry matter.
6.17Breeding Strategies for Modifying Lignin Content

in Sorghum for Cellulosic Ethanol Production
and Electrical Energy
Chemical composition of lignocellulosic feedstocks is a key factor affecting effi-

ciency of cellulosic ethanol production during the biomass conversion process.
Biomass chemical composition and structural characteristics can be affected by
several factors including plant genetics, growth environment, developmental stage,
harvesting method, storage, and others. Since these sources of variation are dif-
ficult to control; tailoring of feedstock chemical composition can be used to attain
high efficiency and optimal biomass conversion.
Currently in Brazil, the abundant and low cost sugarcane bagasse is used as
burning fuel in boilers to produce electricity, making sugarcane mills energy self-
sufficient. However, a considerable amount of bagasse is still not utilized, result-
ing in a waste problem. Therefore, production of cellulosic ethanol could increase
ethanol production in Brazil by up to twofold and also eliminate an environmental
waste problem.
Cellulosic ethanol production is achieved by hydrolyzing cellulose and hemi-
cellulose fractions of biomass in order to release fermentable sugars. There are
several obstacles to biomass conversion that makes developing an economically
feasible technology very difficult (Lynd et al. 2005; Vermerris 2011). In addi-
tion to cellulose recalcitrance, the presence of lignin, an important component of
cell walls, is one of the major problems in cellulosic ethanol production. It has
been shown that biomass conversion yields were highly improved by reducing
lignin content in sorghum lines by carrying the bmr mutations (Dien et al. 2009).
However, in the case of using bagasse as feedstock for water vapor generation,
varieties with higher lignin content are desirable since their biomass have a higher
caloric value.
Therefore, it is important to consider the structural and biochemical charac-
teristics that can be enhanced in energy crops, allowing achievement of set goals
for liquid biofuels production. Ideal energy crops will maximize yield per hec-
tare with minimum inputs and exhibit value-added traits that enhance their use as
biofuel feedstock. In addition to enabling agronomic traits, many argue that new
energy crops must focus on cell wall production since the bulk of photosynthetic
free energy is found in the polymers of this complex matrix (U.S. DOE 2006).
At Embrapa, our objective in developing biomass sorghums is to tailor the bio-
mass composition to increase efficiency of biomass conversion. Sorghum with a
higher level of lignin is desirable for burning, and with a lower level of lignin is
desirable for biomass hydrolysis to fermentable sugars. There are already several
techniques to utilizing sweet sorghum bagasse for burning, as well as methane
and ethanol conversion (Gnansounou et al. 2005; Rooney et al. 2007; Xin and
Wang 2011).
Embrapa Maize and Sorghum is developing very high yielding (5060tha-
1 dry matter) high energy photosensitive hybrids as a feedstock for cellulosic
ethanol. Some of these hybrids have bmr (brown midrib) mutant genes intro-
gressed in different genetic backgrounds, which allows for selection of mate-
rials showing low lignin content without lodging problems. Several of these
hybrids are already being evaluated under different field conditions. Strategies
for studying biomass composition and its accumulation differences, as well as
sugar content and juiciness, are under way and are great resource for high res-
olution mapping of desirable QTLs and/or genes. This is being carried out in
a PS sweet sorghum RIL population (275 F7 individuals) using the genotyp-
ing by sequencing (GBS) method (Elshire et al. 2011). We expect this molecu-
lar approach to significantly contribute to the improvement of sorghum as an
energy dedicated crop.
In a similar approach, Embrapa has also developed a diverse sorghum asso-
ciation panel composed of approximately 200 photosensitive and insensitive
materials (high biomass). The panel has been fully characterized for biomass
compositional traits, including lignin content, using Near-infrared spectroscopy
technology and biochemical assays. This association panel was put together using
accessions from Embrapas germplasm collection and breeding program, as well
as accessions from the CIRAD core collection. The data showed great variability
of cell wall composition among the genotypes. All individuals of this panel will
be sequenced by GBS, allowing us to genotype the panel with thousands of SNPs
across the sorghum genome. This will give us an opportunity for genome-wide
association studies on traits related to biomass quality and yield. Genotyping this
panel using the GBS method will also enable genetic diversity characterization of
the panel, which will be useful in the selection of materials and development of
new mapping populations for biomass accumulation and composition studies in
sorghum.
Together, these different approaches will contribute to the development of sor-

ghum materials with better biomass conversion properties, adding an aggregated
value for producing bagasse/biomass that produce ethanol or electricity.
References
Anderson IC, Buxton DB, Hallam A, Hunter E (1995) Biomass production and ethanol potential
from sweet sorghum. Leopold Center for Sust Agric, Iowa State Univ Ames, IA 4:97101
Barbanti L, Grandi S, Vecchi A, Venturi G (2006) Sweet and fiber sorghum (Sorghum bicolor
(L) Moench), energy crops in the frame of environmental protection from excessive nitrogen
loads. Europ J Agron 25:3039
Bean B, Bronson KF, Schwartz R, Malapati A (2008) Nitrogen requirements of sorghums for
biofuel feedstock production in the southern high plains. In: Proceedings of great plains soil
fertility conference, Denver, CO. 45 March 2008. Intern Plant Nutrition Inst, Norcross, GA,
pp 1518
Bennett AC, Anex RP (2009) Production, transportation and milling costs of sweet sorghum as a
feedstock for centralized bioethanol production in the upper Midwest. Bioresources Technol
100:15951607
Bhattacharya A, Rice N, Shapter FM, Norton SL, Henry RJ (2011) Sorghum. In: Kole C (ed)
Wild crop relatives: genomic and breeding resources (cereals). Springer, Berlin, pp 397406
Boivin K, Deu M, Rami JF, Trouche G, Hamon P (1999) Towards a saturated sorghum map using
RFLP and AFLP markers. Theor Appl Genet 98:320328
Bowers JE, Abbey C, Anderson S, Chang C, Draye X, Hoppe AH, Jessup R, Lemke C,
Lennington J, Li Z, Lin Y-R, Liu S-C, Luo L, Marler BS, Ming R, Mitchell SE, Qiang D,
Reischmann K, Schulze SR, Skinner DN, Wang Y-W, Kresovich S, Schertz KF, Paterson AH
(2003) A high-density genetic recombination map of sequence-tagged sites for Sorghum, as
a framework for comparative structural and evolutionary genomics of tropical grains and
grasses. Genetics 165:367386
Broadhead DM, Freeman KC (1980) Stalk and sugar yield of sweet sorghum as affected by spac-
ing. Agron J 72:523524
Buchanan C, Lim S, Salzman RA, Kagiampakis I, Klein RR, Pratt LH, Cordonnier-Pratt M-M,
Klein PE, Mullet JE (2005) Sorghum bicolors transcriptome response to dehydration, high
salinity and ABA. Plant Mol Biol 58:699720
Carter PR, Barnett KH (1987) Corn-hybrid performance under conventional and no-tillage sys-
tems after thinning. Agronomy J 79:919926
Casa AM, Mitchell SE, Hamblin MT, Sun H, Bowers JE, Paterson AH, Aquadro CF, Kresovich S
(2005) Diversity and selection in sorghum: concurrent studies with simple sequence repeats
(SSRs). Theor Appl Genet 111:2330
Chiaramonti D, Grassi G, Nardi A, Grimm HP (2004) ECHI-T: large bio-ethanol project from
sweet sorghum in China and Italy. Energia Trasporti Agricoltura, Florence
DeHaan LR, Van Tassel DL, Cox TS (2005) Perennial grain crops: a synthesis of ecology and
plant breeding. Renewable Agric Food Syst 20:514
Dien BS, Sarath G, Pedersen JF, Sattler SE, Chen H, Funnell-Harris DL, Nichols NN, Cotta MA
(2009) Improved sugar conversion and ethanol yield for forage sorghum (Sorghum bicolor
L. Moench) lines with reduced lignin contents. BioEnerg Res 2:153164
Dje Y, Heuertz M, Lefebvre C, Vekemans X (2000) Assessment of genetic diversity within and
among germplasm accessions in cultivated sorghum using microsatellite markers. Theor Appl
Genet 100:918925
Doggett H (1988) Sorghum, 2nd edn. Wiley, New York
Dweikat I (2005) A diploid, interspecific, fertile hybrid from cultivated sorghum, Sorghum
bicolor, and the common Johnsongrass weed Sorghum halepense. Molec Breed 16:93101
Dweikat I, Weil CF, Moose SP, Kochian L, Mosier NS, Ileleji KE, Brown PJ, Peer WA, Murphy
AS, Taheripour F, McCann MC, Carpita NC (2012) Envisioning the transition to a next-
generation biofuels industry in the midwest. Biofuels Bioprod Bioref 6:376386
Edwards GE, Franceschi VR, Voznesenskaya EV (2004) Single-cell C4 photosynthesis versus the
dual-cell (Kranz) paradigm. Annu Rev Plant Biol 55:173196
EIA (2011) Annual energy outlook 2011 with projections to 2035. Energy Information
Administration, Washington, D.C
Elshire RJ, Glaubitz JC, Sun Q, Poland JA, Kawamoto K, Buckler ES, Mitchell SE (2011) A
robust, simple genotyping-by-sequencing (GBS) approach for high diversity species. PLoS
ONE 6:e19379. doi:10.1371/journal.pone.0019379
Epplin FM (1996) Cost to produce and deliver switchgrass biomass to an ethanol conversion
facility in the southern plains of the U.S. Biomass Bioenergy 11:459467
FAO (2011) http://faostat.fao.org/site/339/default.aspx (last time accessed: 06/28/2013)
Ferraris R, Charles-Edwards DA (1986) A comparative analysis of the growth of sweet and for-
age sorghum crops II. Accumulation of soluble carbohydrates and nitrogen. Aust J Agric Res
37:513552
Geng S, Hills FJ, Johanson SS, Sah RN (1989) Potential yields and on-farm ethanol cost of corn,
sweet sorghum, fodder beet and sugarbeet. J Agron Crop Sci 162:2129
Glover J (2005) The necessity and possibility of perennial grain crops. Renewable Agric Food
Syst 20:14
Gnansounou E, Dauriat A, Wyman CE (2005) Refining sweet sorghum to ethanol and sugar: eco-
nomic trade-offs in the context of North China. Bioresour Technol 96:9851002
Goodstein DM, Shu S, Howson R, Neupane R, Hayes RD, Fazo J, Mitros T, Dirks W, Hellsten
U, Putnam N, Rokhsar DS (2012) Phytozome: a comparative platform for green plant
genomics. Nucleic Acids Res 40:D1178D1186
Graham R, English B, Noon C (2000) GIS-based modeling system for evaluating the cost of
delivered energy crop feedstock. Biomass Bioenergy 18:309329
Graven LM, Carter PR (1991) Seed quality effect on corn performance under conventional and
no-tillage systems. J Prod Agric 4:366373
Gunaratna N (2002) Early season cold tolerance in sorghum. MS thesis, Purdue University, West
Lafayette, IN
Haussmann BIG, Hess DE, Seetharama N, Welz HG, Geiger HH (2002) Construction of a com-
bined sorghum linkage map from two recombinant inbred populations using AFLP, SSR,
RFLP, and RAPD markers, and comparison with other sorghum maps. Theor Appl Genet
105:629637
Hennigh D, Al-Khatib K, Tuinstra M (2010) Response of acetolactate synthase-resistant grain
sorghum to nicosulfuron plus rimsulfuron. Weed Technol 24:411415
Henry R (2010) Evaluation of plant biomass resources available for replacement of fossil oil.
Plant Biotechnol J 8:288293
Herbst BK (2003) The feasibility of ethanol production in Texas. MS Thesis, Texas A&M
University
Hons FM, Moresco RF, Wiedenfeld RP, Cothren JT (1986) Applied nitrogen and phosphorus
effects on yield and nutrient uptake by high-energy sorghum produced for grain and biomass.
Agron J 76:10101069
Hopkinson CS Jr, Day JW Jr (1980) Net energy analysis of alcohol production from sugarcane.
Science 207:302304
Hunter EL (1994) Development, sugar yield, and ethanol potential of sweet sorghum. MS thesis,
Iowa State University, Ames, Iowa
Hunter EL, Anderson IC (1997) Sweet sorghum. Hortic Rev 21:73104
Jackson DR, Arthur MF, Davis M, Kresovich S, Lawhon WT, Lipinsky ES, Price M, Rudolph
A (1980) Research report on development of sweet sorghum as an energy crop. Volume 1:
Agricultural task
Kim J-S, Klein PE, Klein RR, Price HJ, Mullet JE, Stelly DM (2005) Chromosome identification
and nomenclature of Sorghum bicolor. Genetics 169:11691173
Klein PE, Klein RR, Cartinhour SW, Ulanch PE, Dong J, Obert JA, Morishige DT, Schlueter SD,
Childs KL, Ale M, Mullet JE (2000) A high-throughput AFLP-based method for constructing
integrated genetic and physical maps: progress toward a sorghum genome map. Genome Res
10:789807
Kuepper G (1992) Sweet sorghum: production and processing. Kerr Center for Sustainable
Agriculture, Poteau
Lewis NS, Nocera DG (2006) Powering the planet: chemical challenges in solar energy utiliza-
tion. Proc Natl Acad Sci USA 103:1572915735
Liska AJ, Cassman KG (2008) Towards standardization of life-cycle metrics for biofuels: green-
house gas emissions mitigation and net energy yield. J Bio-based Mater Bioenergy 2:187203
Lueschen WE, Putnam BK, Kanne BK, Hoverstad TR (1991) Agronomic practices for produc-
tion of ethanol from sweet sorghum. J Prod Agric 4:619625
Lynd LR, van Zyl WH, McBride JE, Laser M (2005) Consolidated bioprocessing of cellulosic
biomass: an update. Curr Opin Biotechnol 16:577583
Mapemba LD, Epplin FM, Taliaferro CM, Huhnke RL (2007) Biorefinery feedstock production
on conservation reserve program land. Rev Agric Econ 29:227246
Marguerat S, Bahler J (2010) RNA-seq: from technology to biology. Cell Mol Life Sci
67:569579
Martin PM, Kelleher FM (1984) Effects of row spacing and plant population on sweet sorghum
yield. Aust J Exp Agric Husb 24:386390
Menz MA, Klein RR, Mullet JE, Obert JA, Unruh NC, Klein PE (2002) A high-density genetic
map of Sorghum bicolor (L.) Moench based on 2926 AFLP, RFLP and SSR markers. Plant
Mol Biol 48:483499
Menz MA, Klein RR, Unruh NC, Rooney WL, Klein PE, Mullet JE (2004) Genetic diversity
of public inbreds of sorghum determined by mapped AFLP and SSR markers. Crop Sci
44:12361244
Miller FR, McBee GG (1993) Genetics and management of physiologic systems of sorghum for
biomass production. Biomass Bioenergy 5:4149
Murray D (2005) Ethanols potential: looking beyond corn. Earth Policy Institute, Washington,
D.C. http://www.xmarks.com/site/www.earth-policy.org/Updates/2005/Update49.htm. Last
accessed 06/28/2013
Murray SC, Rooney WL, Hamblin MT, Mitchell SE, Kresovich S (2009) Sweet sorghum genetic
diversity and association mapping for brix and height. Plant Genome 2:4862
Outlaw JL, Ribera LA, Richardson JW, da Silva J, Bryant H, Klose SL (2007) Economics of
sugar-based ethanol production and related policy issues. J Agric Appl Econ 39:357363
Paterson AH, Lin YR, Li Z, Schertz KF, Doebley JF, Pinson SRM, Liu SC, Stansel JW, Irvine JE
(1995) Convergent domestication of cereal crops by independent mutations at corresponding
genetic loci. Science 269:17141718
Paterson AH, Bowers JE, Bruggmann R, Dubchak I, Grimwood J, Gundlach H, Haberer G,
Hellsten U, Mitros T, Poliakov A, Schmutz J, Spannagl M, Tang H, Wang X, Wicker T,
Bharti AK, Chapman J, Feltus FA, Gowik U, Grigoriev IV, Lyons E, Maher CA, Martis
M, Narechania A, Otillar RP, Penning BW, Salamov AA, Wang Y, Zhang L, Carpita NC,
Freeling M, Gingle AR, Hash CT, Keller B, Klein P, Kresovich S, McCann MC, Ming R,
Peterson DG, Mehboob-ur R, Ware D, Westhoff P, Mayer KF, Messing J, Rokhsar DS (2009)
The Sorghum bicolor genome and the diversification of grasses. Nature 457:551556
Pendleton J, Brown CM, Weibel RO (1965) Effect of reflected light on small grain yields. Crop
Sci 5:373
Peng Y, Schertz KF, Cartinhour S, Hart GE (1999) Comparative genome mapping of Sorghum
bicolor (L.) Moench using an RFLP map constructed in a population of recombinant inbred
lines. Plant Breed 118:225235
Perlack RD, Wright LL, Turhollow AF, Graham RL, Stokes BJ, Erbach DC (2005) Biomass
as feedstock for bioenergy and bioproducts industry: the technical feasibility of a billion-
ton annual supply. DOE/GO-102995-2135, ORNL/TM-2005/66. Oak Ridge National
Laboratory, Oak Ridge, Tenn. http://feedstockreview.ornl.gov/pdf/billion_ton_vision.pdf.
Last accessed: 06/28/2013
Pinthus MJ, Rosenblum J (1961) Germination and seedling emergence of sorghum at low tem-
peratures. Crop Sci 1:293296
Piper JK, Kulakow PA (1994) Seed yield and biomass allocation in Sorghum bicolor and F1 and
backcross generations of S. bicolor x S. halepense hybrids. Can J Bot 72:468474
Porter KS, Axtell JD, Lechtenberg VL, Colenbrandu VF (1978) Phenotype fiber composition and
in vitro dry matter disappearance of chemically induced brown-midrib (bmr) mutants of sor-
ghum. Crop Sci 18:205208
Pratt LH, Liang C, Shah M, Sun F, Wang HM, Reid SP, Gingle AR, Paterson AH, Wing R,
Dean R, Klein R, Nguyen HT, Ma HM, Zhao X, Morishige DT, Mullet JE, Cordonnier-Pratt
MM (2005) Sorghum expressed sequence tags identify signature genes for drought, pathogen-
esis, and skotomorphogenesis from a milestone set of 16,801 unique transcripts. Plant Physiol
139:869884
Propheter JL, Wang D, Wu X, Staggenborg SA (2010) Performance of annual and perennial bio-
fuel crops: yield during the first two years. Agron J 102:806814
Reddy BVS, Kumar AA, Ramesh S (2007) Sweet sorghum: a water saving bio-energy crop. In:
International conference on linkages between energy and water management for agriculture
in developing countries, IWMI, ICRISAT, Hyderabad, India
Reddy BVS, Layaoen H, Dar WD, Rao PS, Eusebio JE (2011) Sweet sorghum in the Philippines:
status and future. ICRISAT, Patancheru 502 324, Andhra Pradesh, India
Renewable Energy World (2000) Bioethanol-industrial world perspective. Website www.jxj.com/
magsandj/rew/200_03/bioethanol.html
Ribera LA, Outlaw JL, Richardson JW, Silva JD, Bryant H (2007) Mitigating the fuel and feed
effects of increased ethanol production utilizing sugarcane. St. Louis, MO, Farm Foundation
Conference
Ricaud R, Arceneaux A (1990) Sweet sorghum research on biomass and sugar production in
1990. Manuscript report from the St. Gabriel Experiment Station
Ricaud RB, Cochran B, Arceneaux A, Newton G (1979) Sweet sorghum for sugar and biomass
production in Louisiana. Manuscript report from the St. Gabriel Experiment Station
Rooney WL, Blumenthal J, Bean B, Mullet JE (2007) Designing sorghum as a dedicated bioen-
ergy feedstock. Biofuels, Bioprod Biorefin 1:147157
Salassi ME (2007) The economic feasibility of ethanol production from sugar crops. Louisiana
Agric 50:1
Salzman RA, Brady JA, Finlayson SA, Buchanan CD, Sun F, Klein PE, Klein RR, Pratt LH,
Cordonnier-Pratt M-M, Mullet JE (2005) Transcriptional profiling of sorghum induced by
methyl jasmonate, salicylic acid, and aminocyclopropane carboxylic acid reveals cooperative
regulation and novel gene responses. Plant Physiol 138:352368
Schaffert RE, Borgonovi RA, Casela CR (1986) Metodologia usada para desenvolver novas cul-
tivares de sorgo sacarino. Relatorio Tecnico Anual do Centro Nacional de Pesquisa de Milho
e Sorgo 19801984. Sete Lagoas, Minas Gerais, EMBRAPA/CNPMS, Brazil
Shapouri H, Salassi M, Fairbanks JN (2006) The economic feasibility of ethanol production from
sugar in the United States. USDA-OEPNA and Louisiana State University
Singh SP (1985) Sources of cold tolerance in grain sorghum. Can J Plant Sci 65:251257
Smith GA, Buxton DR (1993) Temperate zone sweet sorghum ethanol production potential.
Bioresour Technol 43:7175
Smith MS, Frye WW, Varco JJ (1987) Legume winter cover crops. Adv Soil Sci 7:95139
Soltani A, Almodares A (1994) Evaluation of the investments in sugar beet and sweet sorghum
production. In: National convention of sugar production from agriculture products, 1316
Mar 1994, Shahid Chamran University, Ahwaz, Iran
Somerville C (2007) Biofuels. Curr Biol 17:R115R119
Tao YZ, Jordan DR, Henzell RG, Mcintyre CL (1998) Construction of a genetic map in a sor-
ghum RIL population using probes from different sources and its comparison with other sor-
ghum maps. Aust J Agric Res 49:729736
Tuinstra MR, Al-Khatib K (2007) New herbicide tolerance traits in sorghum. In: Proceedings of
the 2007 corn, sorghum, and soybean seed research conference and seed expo, Chicago, IL.
57 Dec 2007. Am Seed Trade Assoc, Alexandria, VA
Tuinstra MR, Soumana S, Al-Khatib K, Kapran I, Toure A, van Ast A, Bastiaans L, Ochanda N,
Salami I, Kayentao M, Dembele S (2009) Efficacy of herbicide seed treatments for control-
ling striga infestation of sorghum. Crop Sci 49:923929
Undersander DJ, Lueschen WE, Smith LH, Kaminski AR, Doll JD, Kelling KA, Oplinger ES (1990)
Alternative field crops manual: sorghum syrup. http://www.hort.purdue.edu/newcrop/afcm/
syrup.html. Last accessed: 06/28/2013
US DOE (2006) Breaking the biological barriers to cellulosic ethanol: a joint research agenda.
Report from the December 2005 workshop, DOE/SC-0095. U.S. Department of Energy Office
of Science. http://www.genomicscience.energy.gov/biofuels/. Last accessed: 06/28/2013
Vermerris W (2011) Survey of genomics approaches to improve bioenergy traits in maize, sor-
ghum and sugarcane. J Integr Plant Biol 53:105119
Vermerris W, Saballos A, Ejeta G, Mosier NS, Ladisch MR, Carpita NC (2007) Molecular breed-
ing to enhance ethanol production from corn and sorghum stover. Crop Sci 47:S142S153
Vietor DM, Miller FR (1990) Assimilation, partitioning and nonstructural carbohydrates in sweet
compared with grain sorghum. Crop Sci 30:11091115
Wiendenfeld R (1984) Nutrient requirement and use efficiency by sweet sorghum. Energy Agr
3:4959
Worley JW, Cundiff JS (1991) System analysis of sweet sorghum harvest for ethanol production
in the Piedmont. Trans Amer Soc Agr Eng 34:539547
Wu X, Staggenborg S, Propheter JL, Rooney WL, Yu J, Wang D (2008) Features and fermenta-
tion performance of sweet sorghum juice after harvest. ASABE Paper No. 080037. ASABE,
St Joseph, MI
Xin Z, Wang ML (2011) Sorghum as a versatile feedstock for bioenergy production. Biofuels
2:577588
Xin Z, Wang ML, Barkley NA, Burow G, Franks C, Pederson G, Burke J (2008) Applying gen-
otyping (TILLING) and phenotyping analyses to elucidate gene function in a chemically
induced sorghum mutant population. BMC Plant Biol 8:103
Zhao YL, Dolat A, Steinberger Y, Wang X, Osman A, Xie GH (2009) Biomass yield and changes
in chemical composition of sweet sorghum cultivars grown for biofuel. Field Crops Res
111:5564
Chapter 7
Genetics, Genomics and Crop Modelling:
Integrative Approaches to the Improvement
of Biomass Willows
Angela Karp, Goetz M. Richter, Ian F. Shield and Steven J. Hanley
Abstract Willows (Salix spp.) grown as short rotation coppice (SRC) are among
the leading commercially grown biomass crops in temperate regions, however,
compared with major arable crops they are relatively undomesticated. Initial
advances in improving the crop were made by selecting stem characteristics
(height, diameter, straightness) and coppicing response (shoot number, shoot vig-
our), as well as resistance to pests, diseases and environmental stress. Selections
were achieved purely on the basis of phenotype, with little understanding of the
genetics for many of these important traits, or how they interact with each other
and the environment. To enhance yields further, and to adapt the crop to future cli-
mates and more marginal environments where biomass crops will be encouraged, a
more holistic understanding is needed of the key traits to target and expected gene-
environment interactions. In this chapter we begin by reviewing what is known
about growth in willow in relation to the parameterisation of process-based models
and the advances made in willow genetics and genomics. We finish by considering
an integrative approach which feeds genotypic information into phenotypic models
of source-sink interaction to identify target traits for crop improvement.
Keywords Willow Salix Biomass Growth Genetics Process models

Breeding
A. Karp(*) I. F. Shield S. J. Hanley

AgroEcology Department, Rothamsted Research,
Harpenden AL5 2JQ, Herts, UK
e-mail: angela.karp@rothamsted.ac.uk
G. M. Richter
Dept. of Sustainable Soils and Grassland Systems, Rothamsted Research,
Harpenden AL5 2JQ, Herts, UK
108 A. Karp et al.
7.1Introduction
Willows are catkin-bearing trees of the genus Salix. They occupy a wide variety of eco-
logical niches in temperate zones and are extremely variable in growth characteristics.
An exception is the enclosing of buds within a single scale; one of the characteristics
that distinguish Salix from the related genus Populus (poplars). The circa 400 species
are broadly grouped into the tree willows (sub-genus Salix), the dwarf and alpine wil-
lows (sub genus Chamaetia) and the shrubby willows (sub-genus Vetrix). The latter are
particularly suited as biomass crops due to their propensity for fast, vigorous growth in
coppicing cycles, ease of vegetative propagation and low fertilisation requirements (an
average of 2030kgNha1year1 depending upon site) (Karp et al. 2010a).
Systematic breeding of willow for biomass has only been pursued since the late
1980s in response to a rise in the oil prices, and then in just a few places world-
wide. The potential for further crop improvement is therefore huge. Central to real-
ising this potential are the answers to two inter-related questions: what is growth
in willow and what determines willow growth? These questions are being tackled
indirectly and directly by many groups, including the BSBEC-BioMASS consor-
tium, which is part of the UK BBSRC Sustainable Bioenergy Centre (BSBEC).
BSBEC-BioMASS planted a trial of four different genotypes (Tora, Endurance,
Resolution and Terra Nova) at two sites in the UK (Cerasuolo et al. 2013). A
large number of non-destructive and destructive measurements were made on this
trial, which, together with studies of willow mapping populations, have led to
an improved understanding of many aspects of willow growth. In this review we
draw on these results, and other studies, and show that the answers to the questions
posed above lie in understanding how the plant integrates across many processes.
7.2Growth in Willow
Growth is as an increase in size due to cell division and subsequent cell expan-
sion. Growth processes have been most thoroughly studied in the model plant
Arabidopsis thaliana (Thale Cress) and knowledge is accumulating in pop-
lar as a model tree. We start by considering willow growth in comparison with
Arabidopsis and poplar, as differences become relevant later when processes and
candidate genes derived from these model plants are considered.
7.2.1Perenniality and Seasonal Growth
Arabidopsis thaliana is a self-fertilising hermaphrodite annual. The primary shoot

apical meristem grows monopodially, remaining active throughout the life span of
the plant and continuously producing new lateral organs. Growth is primary and
vegetative until environmental cues (Sect. 7.3.1) result in a developmental switch
to flowering. An inflorescence is produced, seed is set and the plant dies. New
plants develop from the seed.
7 Genetics, Genomics and Crop Modelling 109
Fig.7.1The perennial
cycle of willow and poplar
depicting the main growth
transitions. Although some
stages overlap, or occur
concurrently, they involve
separate developmental
processes and have been
shown distinctly for this
reason
Willows and poplars are obligate (diocecious) outcrossing perennials. They

show sympodial primary and secondary growth. There is a distinct period
in the year, from early spring to late summer when they are actively grow-
ing and the remainder when growth has ceased (Fig.7.1). Growth cessation
marks the beginning of several developmental phases which include senes-
cence, leaf fall, the laying down of reserves, induction of cold hardiness and,
like Arabidopsis, a developmental shift from leaf to bud formation. Unlike
A. thaliana, however, willows and poplars form floral and vegetative buds in
different positions on the stem and the end of the season is not demarked by
flowering and seed production but by bud set and senescence. In poplar a ter-
minal bud is formed at the shoot apex but in willow the shoot tip degenerates,
or even abscises. The plants over-winter in a state of dormancy. New spring
growth begins with the activation of vegetative buds (bud break or bud flush)
to form leaves and floral buds (flowering) to form catkins. In willow there is
considerable variation in catkin morphology and whether they appear before
(precociouse.g. S. daphnoides and S. viminalis), at the same time (coetane-
ous, e.g. S. alba and S. babylonica), or after leaves form (serotinous, e.g.
S. pentandra and S. triandra). Bud break is closely followed by growth
from the new apical meristem in the stem tip and the cambium (Sect. 7.2.3).
In addition, adventitious buds may arise during the growing season to form
sylleptic branches. In poplars these contribute to biomass yield, but not in wil-
low (Sennerby-Forsse 1995).
Flowering is delayed in poplar (for around seven years) but most willows
flower in the first or second year of growth from a seed or vegetative cut-
ting, an important advantage for genetics and breeding. The tiny seed has lit-
tle endosperm and needs to germinate immediately. Fast seedling growth rates
(1.01.4g per week), which exceed standard woody plants, have been reported
(Grime et al. 1988). The seed comprises genetically distinct individuals and is
difficult to handle. Consequently, commercial biomass willows are planted as
stem cuttings of ~20cm length. Growth rates from cuttings are also high, for
example 1.20g per week for S. viminalis under continuous light (MacDonald
1989).
110 A. Karp et al.
7.2.2Coppicing: The Stem and the Stool
In short rotation coppice (SRC), stem cuttings are planted in spring and grown
for one year, when they can reach 23m in height. After leaf fall the stems are
cut back to induce a coppicing response, in which multiple shoots re-sprout from
the cut base (stool) in the following spring. Further growth is allowed, typically
for three more years, by which time the stems are ~7m tall. Specialist machin-
ery is used to harvest the stems to produce chips or billets for storage and use.
Resprouting follows in spring and the cycle is continued for circa 20years.
The SRC cycle has interesting features in terms of growth (Fig.7.2). The stem
cutting has everything required to re-grow a tree: vegetative buds to form leaves,
adventitious nodes/lenticels to develop roots and reserves for initial growth
(Sect. 7.3.3). Growth is similar to that from the intact stem in inter-harvest years.
New growth from cut stools, however, relies on stool and root reserves. It originates
from axillary buds in the stool that may have been dormant for successive years
until stem removal at harvest releases them from apical dominance. The stool starts
as a small swelling and increases in size with each successive harvest. The number
of re-sprouting shoots differs among species (Sennerby-Forsse and Zsuffa 1995).
Axillary buds contain three primordia, each of which has a fixed number of initia-
tion nodes from which leaves develop to give a flush of fixed growth. Additional
free growth arises from new nodes initiated during the year (Sennerby-Forsse
et al. 1984). Many buds sprout simultaneously but the resultant stems are pro-
gressively thinned, a process involving differential growth rates, dominance and
Fig.7.2The main growth characteristics of the short rotation coppice cycle in willow. Events
are shown for the establishment year through the first rotation, including initial re-growth after
the first harvest
suppression of stems and re-allocation of resources (Sennerby-Forsse et al. 1984).

Coppicing re-invigorates growth and its timing is a compromise between the
diminishing yield gained per successive year, harvesting costs, and the need to
replenish sufficient reserves for vigorous re-sprouting (Verwijst 1996a, b).
Root growth in S. viminalis is initiated at least one week before bud break.
Several pulses of increased growth, followed by increased decay occur through-
out the season, suggesting that the roots continually turn-over during shoot growth
(Rytter and Hansson 1996; Rytter 1999, 2001). Root turn-over rate in planted
willow stands was 4.95.8 times year1, and similar to that for willows grown in
lysimeters (4.88.1year1; Rytter and Rytter 1998). In S. triandra (Stott 1961)
and S. dasyclados (Ericsson 1984) 90% of roots occurred within the upper
1220cm. A mean rooting depth of 2530cm was later reported in S. viminalis
(Rytter and Hansson 1996) but roots have been detected at 1.3m. Estimates of the
standing fine root (2mm) biomass, to a depth of 50cm, average 1,320kgha1
(Rytter 1999).
A relationship between maximum root and stem diameters in poplar and willow
accounted for 7.8 and 10.9% of the variation, respectively (Crow and Houston
2004). Older stools showed no evidence of larger roots, suggesting that regular
coppicing may slow down root development or remove the need for larger roots
(Crow and Houston 2004). Rytter found below-ground allocation was highest in
the first year (49 and 58% of total biomass for plants grown in lysimeters on clay
and sand, respectively), and dropped to 3638% (clay) and 3340% (sand) in the
second and third years (Rytter 2001). There is also a general relationship between
stool size and above-ground biomass and both are affected by inter-plant compe-
tition. If planting densities are too high, excessive self-thinning occurs and stool
mortality rates increase as smaller stools are out-competed (Verwijst 1996a, b).
BSBEC-BioMASS has shown that allocation is genotype-dependent. All four gen-
otypes studied showed high biomass yields but Resolution allocated proportionally
more above ground (J. Cunniff, Rothamsted Research UK (unpublished results).
7.2.3Primary and Secondary Growth
The willow shoot tip contains the shoot apical meristem (SAM) and comprises a
stem portion with overlapping upturned leaves that extend above the tip. Shoot tips
differ in length (e.g. 55, 1530 and 2540mm, for S. viminalis, S. caprea, and their
hybrid, respectively) but are generally very small (Sennerby-Forsse et al. 1984).
Leaf initiation originates in the SAM and is alternate along the stem. The SAM
typically contains an outermost layer, which produces the epidermis, and an inner-
most layer which gives rise to the rest of the shoot. Below this is the zone of central
mother cells, beneath which cell divisions produce vertical cell files that become
the pith and the pro-cambium. A few cm lower the vascular cambium is already vis-
ible as a ring. Primary growth is continuous throughout the season, although growth
rate is affected by temperature and stress (Sennerby-Forsse et al. 1984).
112 A. Karp et al.
Secondary growth, leading to an increase in stem thickness, results from

mitotic activity in the cambium which produce sapwood (or secondary xylem)
to the inside and bark (or secondary phloem) to the outside. In most trees large
cohorts of cells within the cambium undergo divisions over discrete time inter-
vals in the growing season, resulting in growth rings. Willows are diffuse or
semi-ring porous, and growth rings are not always distinct although, the larger
thinner walled vessels formed first (earlywood) can normally be distinguished
from the thicker, smaller latewood produced later (Arihan and Guvenc 2011).
The four genotypes of the BSBEC trial showed similar stem anatomy, although
differences were present in the size of the xylem vessels and in the timing of
cambial activity.
Wood formation is a dynamic process and many trees make Reaction Wood
(RW) to structurally reinforce and redirect growth towards the vertical in
response to mechanical or gravitational stress. In willow and poplar RW com-
prises Tension Wood (TW) on the upper (tension) stem and Opposite Wood
(OW) on the lower stem. TW is often characterised by the formation of a gelati-
nous layer, composed mainly of cellulose, within the fibre cells of the secondary
xylem. TW is problematic for the pulp and paper industry as it reduces the qual-
ity of the wood but the high cellulose content means that sugars can be released
more easily in biomass processing, making it attractive for biofuel production
(Liu and Liu 2010). BSBEC-BioMASS has shown that final glucose yields can
be increased by TW induction without a detrimental impact on biomass yield per
se (Brereton et al. 2011).
7.3Determinants of Willow Growth
7.3.1Environmental Cues: Light and Temperature
In poplar, growth cessation occurs principally in response to a change in day

length from long days (LD) to short days (SD) although light quality and tem-
perature are also important, depending on the latitudinal origin of species
(Rohde et al. 2002; Druart et al. 2007; Olsen 2010). Studies of ecotypes of
S. pentandra L indicate that similar environmental cues operate in willow
(Junttila and Kaurin 1990).
Winter dormancy can be considered as an endogenous repression that is main-
tained, even if conditions become permissive to growth, until a specific cue trig-
gers release (Jimenez et al. 2010). The cue used by many perennial species is
quantitative accumulation of cold exposure, or chilling requirement. Evidence sug-
gests this is also true for willow (Cannell et al. 1987). Moreover, willow shows
genetic variation for this trait (Lennartsson and Ogren 2004; Weih 2009). Flushing
early has been shown to result in increased yields but is not without risk, as once
the buds burst they became frost sensitive (Lennartsson and Ogren 2004).
7.3.2Endogenous Cues: Phytohormones
Cytokinin activity increases abruptly coinciding with floral and leaf bud burst in
willow after which levels fall and remain low throughout the summer (Bowen and
Hoad 1968; Alvim et al. 1976). In contrast, ABA levels increase over the win-
ter following leaf abscission, then decrease before growth initiation in the spring
although a further increase in ABA occurs in July just before growth cessation
(Alvim et al. 1976). No correlations between ABA levels and photoperiodicity
have been detected (Alvim et al. 1979).
Regulation of axillary lateral SAMs is a key way in which plant architecture
is determined. The polar auxin transport stream acts as an inhibitor of axillary
bud outgrowth, an effect opposed by cytokinins and strigolactones (Domagalska
and Leyser 2011; Mueller and Leyser 2011). Root apical meristems are con-
trolled negatively by cytokinins and positively by auxin. PIN proteins regulate
auxin flow by localizing it asymmetrically in the vessels. We have demonstrated
that hormone physiology assays developed in Arabidopsis can be directly trans-
ferred to study control of bud activation in biomass willow and that bud hormone
response was qualitatively remarkably similar. Arabidopsis hormone mutants
were used to assess allelic variation in the cognate willow hormone genes and
allelic differences in willow strigolactone genes were observed using this
approach (Ward et al. 2013).
Gibberellins (GAs) are regulators of stem or leaf elongation, and together
with brassinosteroids, are considered to be the main factors influencing plant
height. In poplar cessation of cell division and cell elongation in response to
SD is associated with down regulation of GAs and can be reinitiated by GA
application (Olsen 2010). Similarly, in S. pentandra GAs influence shoot elon-
gation and cessation of apical growth can be prevented by exogenous GAs
(Davies et al. 1985). In elongating shoots of S. dasyclados and S. viminalis
GA1, GA4, GA8, GA9, GA19, GA20 and GA29 were all detected but levels of
GA19 and GA20 were particularly high in both species in the vegetative shoot
(Junttila et al. 1988).
7.3.3Resource Availability and Recycling:

Carbon (C) and Nitrogen (N)
A few weeks prior to any visible new growth in the spring, there is a mobilisa-
tion of resource along the willow stem. The latest formed vessels overwinter in
mature state and are utilised first when growth recommences (Sennerby-Forsse
1986, 1995). Activation of the cambium follows, beginning at the apex and taking
more than a week to spread to the base. Phloem differentiation starts at least two
weeks before flowering and bud break, whilst xylem differentiation occurs around
the same time as bud activation (Sennerby-Forsse 1986).
114 A. Karp et al.
Spring xylem flow comprises a flux of concentrated sucrose solution derived

from the mobilisation of starch reserves in xylem ray cells in the roots and stool.
Sucrose is cleaved near sink tissues to form fructose and glucose by sucrose syn-
thase and invertase. In poplar, sucrose and fructose represent 70 and 90% respec-
tively of the total soluble sugars available to the cambium during active growth
(Deslauriers et al. 2009). As xylem differentiation progresses, the soluble sugars
decrease, indicating that secondary growth rapidly becomes a considerable C sink.
From August onwards, increasing cold hardiness is associated with starch break-
down and an increase in the levels of raffinose and other potential cryoprotectants
(Druart et al. 2007). Parallel changes in flux occur in leaves; in quaking aspen,
sucrose concentrations increase during leaf expansion whilst hexose sugars peak at
the point of mid-expansion and then rapidly decrease (Jeong et al. 2004).
A xylem-to-phloem transfer facilitated by ray cells enables transport of N as
amino acids in the phloem to developing leaves (Cooke et al. 2003). Leaf pho-
tosynthetic capacity shows a strong positive correlation with leaf N content and
vertical leaf N gradient in the canopy is positively correlated with shoot biomass
(Weih and Rnnberg-Wastljung 2007). Resorption of N from leaves and its remo-
bilisation is an important factor in maintaining growth since the more effective this
is the less N is lost from the plant. Pests and diseases, or drought, can result in
premature senescence and reduce photosynthetic capacity too early in the season,
whilst increased N availability can delay growth cessation and senescence and
limit the time for efficient resorption from leaves. Seasonal N cycling is linked
to phenology and is well characterised in poplars (Cooke et al. 2003). Glutamine
and asparagine are the major forms in which N is transported in poplar (Cooke
and Weih 2005). Over winter, N is stored principally in vegetative storage proteins
(VSPs), particularly below the bark (bark storage proteins; BSPs). BSPs show a
characteristic pattern of autumn accumulation and spring disappearance within the
bark, wood and roots. Reactivation of cambial activity in spring coincides with
degradation of BSPs and a rise in amino acid levels (Druart et al. 2007).
7.4Modelling Growth in Willow
Models can increase the selection of improved phenotypes in breeding by iden-

tifying key elements on the basis of their relative importance within the system.
Model-aided crop improvement is a succession of explorations achieved by dis-
secting the processes contributing to productivity into their elements. The number
of processes describing these elements depends on the desired level of complex-
ity and experimental evidence available. For a descriptive simulation of plant
growth, the key components of any model would comprise elements of develop-
ment (phases of phenology), resource capture (leaves, roots) and storage (biomass,
reproductive organs).
A large number of models exist for simulating growth and productivity of trees,
forest or SRC, which can be categorised into research and management tools and
vary greatly in complexity. Models have been specifically developed for SRC wil-
low (Eckersten and Slapokas 1990) and poplar (Deckmyn et al. 2004), of which the
poplar model SECRETS is the most comprehensive and complex, describing a total
of 22 processes and quantitative relationships (Deckmyn et al. 2004). The SECRETS
model was adopted for a similar biological and management system and is thus clos-
est to meeting the requirements needed to describe the trials in BSBEC-BioMASS.
However, it was not considered adequate for our purposes because the carbon (C)
allocation routine originates from a global, upscaled approach, which distinguishes
between green C (leaf, fine roots, etc.) and structural C. This was considered unlikely
to match our specific requirements for a detailed process description of canopy
dynamics, of the interaction between above- and below-ground compartments and
of the effect of reserves, within the sink-source control operating in the willow per-
ennial cycle. To achieve the latter we implemented a sink-source balance described
in LINGRA (Schapendonk et al. 1998) which has been successfully applied in a
generic way to energy grasses (Richter et al. 2010b; Triana et al. 2011). In the fol-
lowing, we outline the conceptual basis of our process-based model and show the
power of the model sensitivity analysis to rank model parameters (Richter et al.
2010a). Ultimately our vision is to link genetic information to the model parameteri-
zation to explain the gene to phenotype (GP) relations (see Sect. 7.6.3).
Our newly developed model for Light Use and Carbon Allocation for Salix
Species (LUCASS) simulates growth and development of SRC willow grown in
monoculture (M. Cerasuolo, Rothamsted Research, UK, unpublished results).
LUCASS considers the processes of phenological and morphological development
of the plant (including senescence and dormancy), light interception, photosyn-
thesis, and respiration, as well as biomass formation. The model defines the plant
organs (buds, leaves, branches, stems, stool and roots) as sinks, to which carbo-
hydrates (CHO) from a common source pool (photosynthates, mobilised reserves)
are allocated. Partitioning follows the principles of balancing demand (sink
strength) and supply (CHO sources). These processes are dependent on tempera-
ture and day length as well as on light and water availability. Boundaries for crop
growth are defined by the soil hydrology and water and energy balance model.
The specificity of our modelling approach was first demonstrated for the simu-
lation of light interception (Cerasuolo et al. 2013) which accounted for the hori-
zontal and vertical structure of the canopy of SRC willow. The model represented
the varietal differences for light interception observed for the four genotypes in the
BSBEC field trial in terms of the distribution of leaf inclination and its effect on
light extinction, but most prominently, in terms of the clumping index and vertical
leaf area distribution. Compared to explicit 3D descriptions of the canopy struc-
ture, this model provides a parsimonious but effective means of identifying geno-
type-specific traits to improve varieties.
The sensitivity analysis applied to LUCASS identified other important vari-
etal traits that characterise light interception, like the leaf shape and leaf extension
rate. The parameters of sink formation; the onset of stem formation (phenological
control related to daylength) and the stem extension rate, proved to be of predomi-
nant importance for yield, across a range of environments. The fraction allocated to
116 A. Karp et al.
aboveground biomass is unsurprisingly the most important parameter, and genotypes

derived from harsh environments would invest more into belowground reserves to
survive periods of growth inhibition. Experimentally, however, this ratio is challeng-
ing to establish, as biomass extraction methods are often incomplete (e.g. ignoring
fine and subsoil roots) and low sampling frequencies do not allow relative allocation
rates to be derived. The experimental setup, such as used in the BSBEC trial, also
does not allow for the quantification of root exudates, which is crucial for estimat-
ing the total carbon allocation balance. Total assimilation rates based on accumulated
biomass will therefore be underestimated, and it will be important to support the cal-
ibration of photosynthesis parameters with instantaneous measurements.
Different process-based models utilise a variable mixture of mechanistic and
empirical understanding that is reflected in the variable number of elements and
mathematical relationships. In developing LUCASS as a first modular step to
developing a full process-based SRC willow model we have identified some
important knowledge gaps and assumptions that need be filled by new experimen-
tation. For the integration of genetic information (Sect. 7.5), the largest challenge
will be to identify and dissect key processes, i.e. to become more mechanistic.
For this, new evidence needs to be generated to bridge the gap between genomic
(QTL, gene) and phenotypic (physiological, pheno-morphological) information
and to calibrate new, dissected process parameters (see Sect. 7.6.3).
7.5Genetic Determinants of Growth Processes

Relevant to Willow
To determine the genetic basis of growth processes affecting yield in willow

key resources for both linkage and association genetics have been developed
and exploited (Karp et al. 2011). One of the largest mapping populations, K8
(n =947), was planted at Long Ashton (Somerset, West UK) in 2000, at RRes
(Hertfordshire, East UK) in 2002 and the RRes farm at Woburn (well-drained,
nutrient-poor soils) in 2009. An additional eleven populations (n=~500) were
also developed at RRes. Several smaller families exist in Sweden and North
America. The Swedish and UK groups also jointly formed an association map-
ping population (n=380) which is planted at contrasting sites in both countries
(Karp et al. 2011).
Using these resources QTLs have been mapped for a large number of traits
in willow including rust resistance (Hanley 2003; Tsarouhas et al. 2003; Hanley
et al. 2011), insect resistance (Rnnberg-Wastljung et al. 2006), shoot height,
stem diameter, and stem number (Tsarouhas et al. 2002), frost tolerance, phenol-
ogy (Tsarouhas et al. 2003, 2004), water-use efficiency and drought tolerance
(Rnnberg-Wastljung et al. 2005; Weih et al. 2006). Willow QTLs are currently
being interrogated with respect to candidate genes from Arabidopsis and pop-
lar. It is not feasible to comprehensively cover the range of genes and gene net-
works that could have relevance here. Instead, an overview is given in Table7.1.
Table7.1Overview of the main genetic controls of growth processes in Arabidopsis and corresponding knowledge in poplar/willow. Only some key genes/
pathways are described. For more comprehensive coverage please see suggested publications in poplar and reviews
Growth process Some key genes in Arabidopsis Simplistic overview in Arabidopsis Key genes/knowledge in pop- References
lar (and/or willow, if known)
Light/Photoperiod PHYTOCHROME (PHY) Photoperiod is perceived in leaves Two PHY genes are known in Frewen et al. (2000),
perception circa- PHYTOCHROME by phytochromes (PHY) poplar: PHYA is involved Ingvarsson et al. (2006),
dian clock INTERACTING FACTORs which entrain components of in dormancy and PHYB Ibanez et al. (2010), Olsen
(PIFs) the circadian clock. Five PHY in bud flush. PHYB2 (2010), Kunihiro et al.
LATE ELONGATED genes are known. They interact co-locates with bud set (2011), Sawa and Kay
HYPOCOTYL1 (LHY1) with phytochrome interacting QTLs on two linkage (2011, 2012)
TIMING OF CAB factors (PIFs). Several genes groups. Single nucleotide
EXPRESSION1 (TOC1) (e.g. LHY1 and TOC1) act as polymorphisms (SNPs) in
FLAVIN-BINDING, KELCH regulatory clock components, PHYB2 showed clinal var-
REPEAT F-BOX1 (FKF1) interacting with other pathways. iation, suggesting it may
7 Genetics, Genomics and Crop Modelling
E.g.. in blue light (daytime) be involved in adaptive

GIGANTEA (GI)
a circadian clock-controlled response to photoperiodic
FKF1-GI complex degrades conditions. Poplar regula-
transcriptional repressors of CO tory clock components
and regulates the timing of CO include PtLHY1, PtLHY2
expression and PtTOC1
(continued)
117
Table7.1continued
118
Growth process Some key genes in Arabidopsis Simplistic overview in Arabidopsis

Key genes/knowledge in pop- References
Vegetative to floral FLOWERING LOCUS C Time measurement in the photo- No structural orthologue Bohlenius et al. (2006), Hsu
transition (at least (FLC) periodic flowering pathway is of FLC is known in et al. (2011, 2012), Sawa
five pathways are CONSTANS (CO) regulated by daytime expres- poplar but there is an FT and Kay (2011), Pin and
known: a key one FLOWERING LOCUS T (FT) sion of CO. CO upregulates FT duplication (FT1 and Nilsson (2012), Kemi et al.
is shown here) bZIP TRANSCRIPTIONAL expression whilst FLC, a strong FT2, on chromosomes (2013)
FACTOR repressor of flowering, directly VIII and X). Control via
FRIGIDA (FRI) represses FT. Under LD CO the CO/FT regulon model
expression coincides with light, has been demonstrated
high CO in leaves upregulates in SD-induced growth
FT, the FT florigen goes to cessation and bud set.
the shoot apex and complexes However, over-expression
with the bZIP transcriptional of CO gave no evidence
factor to activate expression of to support a role of CO
floral-meristem identity genes in regulating bud set or
(below). In SD, light/CO/ bud flush. A pulse of
FT coincidence is lost, FT is FT1 expression in winter
repressed and floral meristems appears to initiate the
do not form. In vernalization- transition from veg-
requiring accessions (e.g. etative to reproductive
perennial A. lyrata) FRI upregu- meristems. FT2 forms
lates FLC whilst prolonged molecular networks
cold (vernalization) overrides with different genes in
FRI enabling flowering when response to various stress
temperatures warm factors to control vegeta-
tive growth
(continued)
A. Karp et al.
Table7.1continued
Growth process Some key genes in Arabidopsis Simplistic overview in Arabidopsis Key genes/knowledge in pop- References
Meristem identity: LEAFY (LFY1) All pathways involved in vegetative Studies of a LF orthologue Fernando and Zhang (2006),
floral primordial APETALA1 (AP1) to reproductive transitions con- (PtLFY;) identified in Srikanth and Schmid
formation verge on meristem identity genes, poplar and AP1 ortho- (2011), Hsu et al. (2012),
of which LFY is the most impor- logue (SAP1-1. from Siriwardana and Lamb
tant. LFY occurs at low levels in S. discolour suggest simi- (2012)
leaf primordia but once above lar controls will operate in
a threshold identity switches to willow to switch meristem
floral primordia. AP1 encodes identity from leaf to floral
a transcription factor with a primordial)
MADS-domain and determines
sepal and petal development
Cell division CYCLINs (CYCs) The cyclin family of proteins regu- 22 CYCD genes were identi- Menges et al. (2007), Dong
CYCLIN-DEPENDENT late the cell cycle by interacting fied in the poplar (Populus et al. (2011)
KINASEs (CDK) with cyclin (dependent kinases trichocarpa) genome; six
CDKs). Arabidopsis contains 10 CYCD subgroups are con-
CYCD genes served across higher plants
Cell wall synthesis CELLULOSE SYNTHASE CesA encode cellulose synthase Many of the cell wall Sarkar et al. (2009), Demura
and secondary (CesA) catalytic subunits. Lignin biosynthesis genes have and Ye (2010), Douglas
growth GLYCOSYLTRANSF ERASES biosynthesis is a branch of the been identified in poplar, et al. ( 2011), Carroll et al.
GTs (IRXs) phenylpropanoid pathway and particularly in the lignin (2012), Li et al. (2012),
VASCULAR-RELATED NAC includes several gene such ass pathway. In addition, a Sanchez-Rodriguez
DOMAIN (VND) PAL, C4H, 4CL, CCR, HCT, number of transcription et al. (2012), Schuetz et al.
C3H, CCoAOMT, CCR, F5H, factors regulate differen- (2013)
COMT, CAD. GTs (IRX genes) tiation of vessels. proteins
are involved in xylan (hemicel- (VND6, VND7, SND1 and
lulose) biosynthesis NST1) are key switches
regulating a cascade of
downstream transcription
119
factors leading to second-

ary wall biosynthesis
(continued)
Table7.1continued
120

Shoot apical meris- KNOTTED1 (KN-1) like KNOX1 homeobox transcription In poplar the role of WUS Schrader et al. (2004), Bao
tems (SAMs) and (KNOX1) factors STM and BP play key and STM in both SAM et al. (2009), Demura and
root apical meris- SHOOTMERISTEMLESS roles in formation of the embry- and axillary meristems Ye (2010), Du and Groover
tem (RAMs) (STM) ogenic SAM. CUC2, NAM has been confirmed but (2010)
BREVIPEDICELLUS (BP) (identified in Petunia) all affect functional orthologues of
CUPSHAPED COTYLEDON SAMs but through different WUS and STM (PttWUS
2 (CUC2) mechanisms. The identity and and PttCLV3) do not
PINHEAD (PNH) maintenance of stem cells in the appear to play a role in
NO APICAL MERISTEM SAM central zone is regulated the cambium. Instead
(NAM) by WUS which interacts with KNOX genes show
CLV3 to regulate the size of the high expression and a
WUSCHEL (WUS)
organizing centre and the stem poplar orthologue of STM
CLAVATA3 (CVL3) cell niches. Mutations stm, cvl1, (ARBORKNOX1;ARK1)
CLAVATA-LIKE19 (CVLE19) cvl3, pnh and wus affect lateral regulates cambial func-
WUS-LIKE (WOX5) and primary SAMs, indicating tions and cell differen-
SCARECROW (SCR) common regulation. A WUS- tiation during secondary
like gene (WOX5) and a CVL3 growth, including regula-
homologue (CLE19) regu- tion of cell wall biosyn-
late the root apical meristem thesis. Similarly, a poplar
(RAM), whilst SCR is involved BP ortholog (ARK2) is
in specifying tissue identity in expressed in both the
RAMs cambium and in lignify-
ing cells
(continued)
A. Karp et al.
Table7.1continued
Shoot and root INDOLE ACETIC ACID Key phytohormone gene networks In poplar over expression Koornneef et al. (1998),
growth, apical (IAA) relate to their biosynthesis, of PcGA2ox resulted in Hartweck (2008),
dominance ABSCISIC ACID (ABI) degradation and signalling. The reduced apical domi- Nieminen et al. (2008),
architecture GIBBERELLINS (GA)) rate limiting step of cytokinin nance, short trees and Domagalska and Leyser
DELLAs biosynthesis is catalysed by wide crowns. Involvement (2011), Mueller and Leyser
ISOPENTYL TRANSFERASE IPTs, degradation is by CKXs of ABI and auxin related (2011)
(IPTs) whilst receptors include the genes has been dem-
AHK family. Auxin synthesis onstrated in relation to
CYTOKININ OXIDASE/
involves IAA whilst signalling drought stress and root
DEHYDROGENASE
includes ARRs and shoot growth and
(CKXs)
GA biosynthesis involves GA auxins, cytokinins and str-
ARABIDOPSIS HISTIDINE
oxidases and signalling includes igolactones in branching

KINASE (AHKs) receptors such as GIDI.
ARABIDOPSIS RESPONSE DELLA proteins also play key
REGULATORS (ARRs) roles in the gibberellin signal-
CAROTENOID CLEAVAGE ling pathway. Strigolactones are
DIOXYGENASE (CCD7) derived from carotenoids, e.g.
via CCD7. The ABI biosyn-
thetic pathway originates in the
chloroplast with hydroxylation
of -carotene to zeaxanthin.
ABI interact with many genes
related to stress, e.g. dehydrins
121
122 A. Karp et al.
Although omic studies are only just beginning in willow, the derivation of the
whole genome sequence of P. trichocarpa (Tuskan 2006) has resulted in data
sets and resources that are highly beneficial (Yang et al. 2009). Transfer of infor-
mation was facilitated by direct alignment of the K8 map to the poplar genome
and the high degree of macrosynteny found between them (Hanley et al. 2006).
More recently, full genome sequencing of willow has been carried out and draft
sequences are close to publication. Transcriptomic and metabolomic studies have
also been conducted, but not yet published. Such approaches can be combined
with QTL analyses, for example to identify e-QTLs or mQTL, respectively (Sect.
7.6), that can be informative in analysis of regulatory pathways. In terms of gene
validation, the disadvantage of willow is that it currently lacks a robust transfor-
mation technology. However, advancements have been made in understanding the
role played by key genes through their study in transgenic Arabidopsis and pop-
lar. Moreover, functional variation of willow alleles can sometimes be assessed by
transforming them into Arabidopsis (Ward et al. 2013) or poplar lines.
7.6An Integrated Understanding of Growth

and Biomass Yield in Willow
Given the complexity of growth process in willow (Sect. 7.2), the plethora of exog-
enous and endogenous cues and gene networks involved (Sects. 7.3 and 7.5), and
the large lists of possible genes (e.g. Table7.1), how is it possible to identify key
targets for breeding or genes to select? There has been a technological revolution in
the way that genomes and their expression can be interrogated in a global systems
way and gene networks affecting traits can be searched for. However, these need to
be combined with hypotheses based on biology and on knowledge of the variation
used by the plants in competing for growth and meeting environmental challenges.
For the use of models describing the interaction between plants and the environment
(Sect. 7.4), new avenues will have to be developed for integrating experimental evi-
dence with genetic information in the parameterization of predictive tools.
7.6.1New Technology-led Approaches
Transcriptomics, metabolomics and/or proteomics can contribute to a systems-

based understanding of a critical developmental transition and/or to compare dif-
ferent developmental stages, tissues or genotypes. Such approaches has been used
to investigate many of the developmental stages of Fig.7.1 in poplar, including
senescence (Andersson et al. 2004), secondary wood formation (Schrader et al.
2004), dormancy and terminal bud formation (Ruttink et al. 2007).
From such studies a more holistic and informed picture can be developed of
how different genes come into play, initially in sensing the environmental cues,
then resulting in changes in meristem identity, development and resource alloca-

tion. For example Ruttink et al. (2007) showed that SD- induced bud formation
was associated with genes involved in light signal transduction and the circadian
clock (e.g. PHYB, TOC1 and CO/FT, GAs and ABA Table7.1). As bud structure
became visible, changes in expression associated with WUS, CLV and KNOX1
pathways became apparent, as well as down regulation of genes involved in the
cell cycle, (e.g. CYCA1 and CDKB). A few weeks after SD induction genes in the
glyoxylate pathway were up-regulated and photosynthetic pathways down-regu-
lated, and carbohydrate metabolism was increasingly modified towards accumula-
tion of storage products and vcryoprotectants.
Due to the excessive numbers of genes whose expression is altered during
these transitions, these approaches, alone, do not easily lead to the identification
of a single critical gene to target in breeding. A transcriptomic study of short day-
induced apical (terminal) bud formation in white spruce (Picea glauca), identified
4,460 differentially expressed sequences (El Kayal et al. 2011), for example. This
was reduced to 108 genes differentially expressed only in developing buds but a
large effort is still needed to identify which of these are important to change with
respect to achieving crop improvement.
A step forward lies in combining omic approaches with genetics, for exam-
ple, by transcriptome analysis of mutants, or genotypes differing in QTL loci/
alleles, or transgenic plants. Expression profiles can also be treated as pheno-
types. In genetical genomics thousands of gene expression levels can be assayed
and the expression phenotypes are subjected to standard QTL analysis to derive
expression QTLs (eQTLs). The eQTL location coincides with that of the regulated
gene in cis-regulation, while trans-acting eQTLs identify regulatory elements
elsewhere in the genome. eQTLs may be evenly spread or appear in hotspots,
depending on the genetic architecture of the gene interactions. Master regu-
lons that underlie trans-eQTL hotspots (hubs) are of particular interest as they
tend to be at the center of gene expression networks (network eQTLs). Similar
approaches can be used to derive mQTL (from metabolite profiling).
Bioinformatics tools have been developed to help mine the information col-
lected in the large data sets that are accumulating from these kinds of omics
studies. However, knowing how to spot something of significance (i.e. defining the
question) for crop improvement remains an issue, especially for complex traits for
which there is little prior knowledge.
7.6.2New Biology-led Approaches
In willow, as we have shown, the biology of growth is well characterised although

knowledge of the determinants of growth has mostly been inferred from model
plants. To target key biology, we have pursued a systematic approach in which we
framed a number of initial hypotheses for crop improvement and built empirical
models based on this available knowledge. In BSBEC-BioMASS the hypotheses
124 A. Karp et al.
were based on: extending the growing season, improving architecture and increas-
ing above-ground allocation. Data from phenotyping the BSBEC trial and map-
ping populations in the field is being used to refine these hypotheses and develop
process-based models to identify critical processes, developmental stages and
component traits. Simultaneously we have been building and deploying genetic
and genomic approaches to map QTL and identify key genes associated with these
targets, using some of the technological approaches outlined above.
Phenotypic data collected for the growth stages outlined in Fig.7.1 have dem-
onstrated that variation exists for the majority of traits examined in willow. This
includes: bud flush, biomass yield (stem heights, diameters and numbers) growth
cessation and senescence, canopy and leaf traits and above and below-ground allo-
cation, as well compositional traits (e.g. lignin, cellulose and hemicellulose con-
tent). However, examination of the way the traits co-locate with respect to yield
QTL and the use of process-based modelling and sensitivity analysis (Sect. 7.4)
have shown that some of the traits which are variable and have a high heritability
are not the most important ones to target in breeding for improved yield.
Efforts to map bud flush (as a way of extending the growing season) have been
successful. A consistent QTL was detected in several mapping populations in the
same location and a possible causal candidate identified, whose function makes
sense with respect to the networks listed in Table7.1. However, integration of
the results with other data from our studies and the results of sensitivity analy-
ses suggests that it is the events prior to bud flush (e.g. the chilling requirement
and resource mobilisation), rather than bud flush per se, that may be more impor-
tant for yield improvement, at least in UK environments. The sensitivity analy-
sis also showed that allocation is a key trait affecting yield, again suggesting that
resource mobilisation events prior to new spring growth and/or dormancy may be
more important than bud flush or bud set per se. Importantly BSBEC-BioMASS
results indicate that there is not a set relationship between above and below-
ground allocations and that selecting for more above-ground will not necessarily
be at the expense of biomass belowground. Experiments in which plants differing
in the bud flush QTL are sampled in the weeks prior to bud flush are now being
performed at RRes and specific experiments have been designed to investigate
resource mobilisation throughout the perennial cycle.
Willows show variation in leaf shape, leaf area and leaf area index (LAI), but plants
with quite different canopy architecture can attain similar high yields. Our process-
modelling, (Sect. 7.4) revealed that key parameters are not leaf area or LAI per se but
the vertical distribution profile of leaf area and clumping index. Efficient light pen-
etration through the canopy can be achieved for willows with large LAI if leaves also
show a high degree of clumping (Cerasuolo et al. 2013). These results provide guide-
lines with which to assess canopy structures associated with different stem numbers.
BSBEC-BioMASS also assessed willow biomass as a feedstock for sugar
release in saccharification tests for biofuel production. Genetic variation in sacchar-
ification potential was assessed and QTL mapped (Brereton et al. 2010). This has
led to a number of potential targets that are now being pursued. However, the most
interesting data came when correlations were sought with biomass composition
and with growth traits, and none were found. However, when a RW phenotype was
induced in eight genotypes grown in pots, glucose release strongly correlated with
the glucose release obtained for mature field-grown trees. No such correlation was
found when RW was not induced. This suggests that genotypic differences in RW
response may be a primary determinant of the variation observed in sugar release
from willow biomass, and could be selected for (Brereton et al. 2012). Experiments
are now underway to identify the genetic basis of the differences in RW response.
7.6.3New Model-led Approaches
The advantages of modular model structure, and the separation of code and param-
eter space, of process-based models, enable the down-scaling (dissection) and
parameterisation of key processes at higher granularity. Hammer et al. (2006) stated,
new models can help in navigating the biological complexity in breeding improved
crop plants, developing predictive tools for the genotypic controls in phenotypic
(GP) models. In their paper, they refer to the combined analyses of QTLs and an
eco-physiological model to address the genetic variability of leaf growth among 100
recombinant inbred lines in response to meristem temperature, leaf water potential,
and vapour pressure deficit (Reymond et al. 2003). Our SRC model has similar rela-
tionships between leaf extension rate and environmental variables, and it would be
interesting to evaluate these for the K8 mapping population in the context of QTL
analysis. Messina et al. (2011) conceptualise an iterative framework to build realis-
tic GP models based on extensive experimentation to establish the relationship
between genomics and phenomics. To follow this, it may be necessary to expand
physiological models implementing modules that represent 3D structural com-
ponents as functional-structural plant models (FSPM; Vos et al. 2010), although a
pseudo-3D model (Cerasuolo et al. 2013) might suffice. The dynamics of resprout-
ing and die-back show similarity to the tillering and tiller suppression observed
in cereals as a function of red:far red ratio, where FSPM proved useful (Vos et al.
2010). Certainly, the sink-source regulation implemented as an up-scaled morpho-
genesis in LUCASS can be down-scaled using the concepts established in an FSPM
like GreenLab (Guo et al. 2006), or other such approaches (Dingkuhn et al. 2007).
Parameters determining onset of sink formation and sink size proved to be of crucial
importance in the sensitivity analysis of LUCASS; however, dynamics of stem num-
ber were modelled empirically, and not as an evidence-based process.
7.7Concluding Remarks
Further improvement of willow as a biomass crop requires an understanding of the

biology of growth and the key target traits that will result in gains in yield and/or
improved environmental performance. We have found that integrating knowledge
126 A. Karp et al.
from phenotyping of specific genotypes in field trials with genetic mapping and
crop modelling in an iterative way has enabled the identification of key component
traits that contribute to useful variation in the field and the developmental stages that
are most critical to this. QTL can then be mapped and advanced omic technolo-
gies applied, as appropriate, to identify markers for breeding. Process-based models
can also be used to help predict yield gains and performance in different environ-
ments. Despite the fact that willows cannot be transformed, this approach has led
us to the identification of likely casual genes and the first results for bud flush, yield
(increase in height and stem diameters) are currently in preparation as publications.
Acknowledgments The authors would like to thank the UK Biotechnological and Biological
Sciences Research Council (BBSRC) and Ceres Inc. for funding support of the BBSRC
Sustainable Bioenergy Centre (BSBEC): Perennial Bioenergy Crops Programme (BB/
G016216/1: BSBEC-BioMASS) and BBSRC for funding the RRes Cropping Carbon Institute
Strategic Programme Grant. We would also like to thank Jennifer Cunniff, Marianna Cerasuolo,
Cristina Gritsch, Tim Barraclough and March Castle (RRes) and Sarah Purdy, Lawrence
Jones and Anne Maddison (IBERS) for their dedicated work as part of the BSBEC-BioMASS
programme and William Macalpine, Rachel Rossiter and Peter Fruen (RRes) for general
scientific support. Rothamsted Research is an Institute supported by the BBSRC.
References
Alvim R, Hewett EW, Saunders PF (1976) Seasonal variation in hormone content of wil-
low.1. Changes in abscisic acid content and cytokinin activity in xylem sap. Plant Physiol
57:474476
Alvim R, Saunders PF, Barros RS (1979) Abscisic acid and the photoperiodic induction of dor-
mancy in Salix viminalis L. Plant Physiol 63:774777
Andersson A, Keskitalo J, Sjodin A, Bhalerao R, Sterky F, Wissel K, Tandre K, Aspeborg H,
Moyle R, Ohmiya Y, Brunner A, Gustafsson P, Karlsson J, Lundeberg J, Nilsson O, Sandberg
G, Strauss S, Sundberg B, Uhlen M, Jansson S, Nilsson P (2004) A transcriptional timetable
of autumn senescence. Genome Biol 5(R24):2113
Arihan O, Guvenc A (2011) Studies of the anatomical structure of stems of willow (Salix L.) spe-
cies (Salicacea) growing in Ankara Province, Turkey. Turk J Bot 35:117
Bao YH, Dharmawardhana P, Mockler TC, Strauss SH (2009) Genome scale transcriptome anal-
ysis of shoot organogenesis in Populus. BMC Plant Biol 9(132)
Bohlenius H, Huang T, Charbonnel-Campaa L, Brunner AM, Jansson S, Strauss SH, Nilsson O
(2006) CO/FT regulatory module controls timing of flowering and seasonal growth cessation
in trees. Science 312:10401043
Bowen MR, Hoad GV (1968) Inhibitor content of phloem and xylem sap obtained from willow
(Salix viminalis L.) entering dormancy. Planta 81:6470
Brereton NJ, Pitre FE, Hanley SJ, Ray M, Karp A, Murphy RJ (2010) Mapping of enzymatic
saccharification in short rotation coppice willow and its independence from biomass yield.
Bioenergy Res 3:251261
Brereton NJB, Pitre FE, Ray MJ, Karp A, Murphy RJ (2011) Investigation of tension wood for-
mation and 2,6-dichlorbenzonitrile application in short rotation coppice willow composition
and enzymatic saccharification. Biotechnol Biofuels 4(83)
Brereton NJ, Ray MJ, Shield I, Martin P, Karp A, Murphy RJ (2012) Reaction wooda key
cause of variation in cell wall recalcitrance in willow. Biotechnol Biofuels 5:83
Cannell MGR, Milne R, Sheppard LJ, Unsworth MH (1987) Radiation interception and produc-
tivity of willow. J Appl Ecol 24:261278
Carroll A, Mansoori N, Li S, Lei L, Vernhettes S, Visser RGF, Somerville C, Gu Y, Trindade LM

(2012) Complexes with mixed primary and secondary cellulose synthases are functional in
Arabidopsis plants. Plant Physiol 160:726737
Cerasuolo M, Richter GM, Cunniff J, Purdy S, Shield I, Karp A (2013) A pseudo-3D model
to optimise the target traits of light interception in short-rotation coppice willow. Agric
Meteorol 173:127138
Cooke JEK, Weih M (2005) Nitrogen storage and seasonal nitrogen cycling in Populus: bridging
molecular physiology and ecophysiology. New Phytol 167:1930
Cooke JEK, Brown KA, Wu R, Davis JM (2003) Gene expression associated with N-induced
shifts in resource allocation in poplar. Plant, Cell Environ 26:757770
Crow P, Houston TJ (2004) The influence of soil and coppice cycle on the rooting habit of short
rotation poplar and willow coppice. Biomass Bioenergy 26:497505
Davies JK, Jensen E, Juntilla O, Rivier L, Crozier A (1985) Identification of endogenous gibber-
ellins from Salix pentandra. Plant Physiol 78:473476
Deckmyn G, Laureysens I, Garcia J, Muys B, Ceulemans R (2004) Poplar growth and yield
in short rotation coppice: model simulations using the process model SECRETS. Biomass
Bioenergy 26:221227
Demura T, Ye Z-H (2010) Regulation of plant biomass production. Curr Opin Plant Biol 13:299304
Deslauriers A, Giovannelli A, Rossi S, Castro G, Fragnelli G, Traversi L (2009) Intra-annual
cambial activity and carbon availability in stem of poplar. Tree Physiol 29:12231235
Dingkuhn M, Luquet D, Clement-Vidal A, Tambour L, Kim HK, Song YH (2007) Is plant growth
driven by sink regulation? In: Spiertz JHJ, Struik PC, Van Laar HH (eds) Scale and complex-
ity in plant systems research: gene-plant-crop relations. Springer, New York, pp 157170
Domagalska MA, Leyser O (2011) Signal integration in the control of shoot branching. Nat Rev:
Mol Cell Biol 12:211221
Dong Q, Zhao Y, Jiang HJ, He H, Zhu S, Cheng B, Xiang Y (2011) Genome-wide identification
and characterization of the cyclin gene family in Populus trichocarpa. Plant Cell Tiss Organ
Cult 107:5567
Douglas CJ, Ehlting J, Harding SA (2011) Phenylpropanoid and phenolic metabolism in
Populus: gene family structure and comparative and functional genomics. In: Joshi CP (ed)
Genetics genomics and breeding of poplar. CRC press, Boca Raton, pp 304326
Druart N, Johansson A, Baba K, Schrader J, Sjodin A, Bhalerao RR, Resman L, Trygg J, Moritz
T, Bhalerao RP (2007) Environmental and hormonal regulation of the activity-dormancy
cycle in the cambial meristem involves stage-specific modulation of transcriptional and meta-
bolic networks. Plant J 50:557573
Du J, Groover A (2010) Transcriptional regulation of secondary growth and wood formation. J
Integr Plant Biol 52:1727
Eckersten H, Slapokas T (1990) Modeling nitrogen turnover and production in an irrigated short-
rotation forest. Agric Meteorol 50:99123
El Kayal W, Allen CCG, Ju CJT, Adams E, King-Jones S, Zaharia LI, Abrams SR, Cooke JEK
(2011) Molecular events of apical bud formation in white spruce, Picea glauca. Plant, Cell
Environ 34:480500
Ericsson T (1984) Root biomass distribution in willow stands grown on bog. In: Perttu K (ed)
Ecology and management of forest biomass. Swed Univ Agric Sci Rep 15:335348
Fernando DD, Zhang SL (2006) Constitutive expression of the SAP1 gene from willow (Salix
discolor) causes early flowering in Arabidopsis thaliana. Devel Genes Evol 216:1928
Frewen BE, Chen THH, Howe GT, Davis J, Rohde A, Boerjan W, Bradshaw HD (2000)
Quantitative trait loci and candidate gene mapping of bud set and bud flush in Populus.
Genetics 154:837845
Grime JP, Hodgson JG, Hunt R (1988) Comparative plant ecology. A functional approach to
common British species. Unwin Hyman, London
Guo Y, Ma YT, Zhan ZG, Li BG, Dingkuhn M, Luquet D, De Reffye P (2006) Parameter opti-
mization and field validation of the functional-structural model GREENLAB for maize. Ann
Bot 97:217230
128 A. Karp et al.
Hammer G, Cooper M, Tardieu F, Welch S, Walsh B, van Eeuwijk F, Chapman S, Podlich D

(2006) Models for navigating biological complexity in breeding improved crop plants.
Trends Plant Sci 11:587593
Hanley SJ (2003) Genetic mapping of important agronomic traits in biomass willow. PhD thesis,
University of Bristol, UK
Hanley SJ, Mallott MD, Karp A (2006) Alignment of a Salix linkage map to the Populus genomic
sequence reveals macrosynteny between willow and poplar genomes. Tree Genet Genom
3:3548
Hanley SJ, Pei MH, Powers SJ, Ruiz C, Mallott MD, Barker JHA, Karp A (2011) Genetic map-
ping of rust resistance loci in biomass willow. Tree Genet Genom 7:597608
Hartweck LM (2008) Gibberellin signaling. Planta 229:113
Hsu C-Y, Adams JP, Kim H, No K, Ma C, Strauss SH, Drnevich J, Vandervelde L, Ellis JD,
Rice BM, Wickett N, Gunter LE, Tuskan GA, Brunner AM, Page GP, Barakat A, Carlson
JE, dePamphilis CW, Luthe DS, Yuceer C (2011) FLOWERING LOCUS T duplication coor-
dinates reproductive and vegetative growth in perennial poplar. Proc Natl Acad Sci USA
108:1075610761
Hsu C-Y, Adams JP, No K, Liang H, Meilan R, Pechanova O, Barakat A, Carlson JE, Page GP,
Yuceer C (2012) Overexpression of constans homologs CO1 and CO2 fails to alter normal
reproductive onset and fall bud set in woody perennial poplar. Plos ONE 7(e45448)
Ibanez C, Kozarewa I, Johansson M, Ogren E, Rohde A, Eriksson ME (2010) Circadian clock
components regulate entry and affect exit of seasonal dormancy as well as winter hardiness
in Populus trees. Plant Physiol 153:18231833
Ingvarsson PK, Garcia MV, Hall D, Luquez V, Jansson S (2006) Clinal variation in phyB2, a can-
didate gene for day-length-induced growth cessation and bud set, across a latitudinal gradient
in European aspen (Populus tremula). Genetics 172:18451853
Jeong ML, Jiang HY, Chen HS, Tsai CJ, Harding SA (2004) Metabolic profiling of the sink-to-
source transition in developing leaves of quaking aspen. Plant Physiol 136:33643375
Jimenez S, Reighard GL, Bielenberg DG (2010) Gene expression of DAM5 and DAM6 is sup-
pressed by chilling temperatures and inversely correlated with bud break rate. Plant Mol Biol
73:157167
Junttila O, Kaurin A (1990) Environmental control of cold acclimation in Salix pentandra. Scand
J For Res 5:195204
Junttila O, Abe H, Pharis RP (1988) Endogenous gibberellins in elongating shoots of clones of
Salix dasyclados and Salix viminalis. Plant Physiol 87:781784
Karp A, Hanley SJ, Trybush SO, Macalpine W, Pei M, Shield I (2011) Genetic improvement of
willow for bioenergy and biofuels. J Integr Biol 53:151165
Kemi U, Niittyvuopio A, Toivainen T, Pasanen A, Quilot-Turion B, Holm K, Lagercrantz U,
Savolainen O, Kuittinen H (2013) Role of vernalization and of duplicated FLOWERING
LOCUS C in the perennial Arabidopsis lyrata. New Phytol 197:323335
Koornneef M, Leon-Kloosterziel KM, Schwartz SH, Zeevaart JAD (1998) The genetic and
molecular dissection of abscisic acid biosynthesis and signal transduction in Arabidopsis.
Plant Physiol Biochem 36:8389
Kunihiro A, Yamashino T, Nakamichi N, Niwa Y, Nakanishi H, Mizuno T (2011) Phytochrome-
Interacting Factor 4 and 5 (PIF4 and PIF5) activate the homeobox ATHB2 and auxin-induc-
ible IAA29 genes in the coincidence mechanism underlying photoperiodic control of plant
growth of Arabidopsis thaliana. Plant Cell Physiol 52:13151329
Lennartsson M, Ogren E (2004) Clonal variation in temperature requirements for budburst and
dehardening in Salix species used for biomass production. Scand J For Res 19:295302
Li S, Lei L, Somerville CR, Gu Y (2012) Cellulose synthase interactive protein 1 (CSI1) links
microtubules and cellulose synthase complexes. Proc Natl Acad Sci USA 109:185190
Liu Y, Liu S (2010) Formation and anatomical characteristics of tension wood in stem of poplar
I-107 seedlings (Populus euramericana cv. 74/76) induced by artifical inclination. Scientia
Silvae Sinicae 46:133140
MacDonald AJS (1989) Nitrate availability and shoot area development in small willow (Salix
viminalis). Plant, Cell Environ 12:417424
Menges M, Pavesi G, Morandini P, Bogre L, Murray JAH (2007) Genomic organization and evo-
lutionary conservation of plant D-Type cyclins(1 C W). Plant Physiol 145:15581576
Messina CD, Podlich D, Dong Z, Samples M, Cooper M (2011) Yield-trait performance land-
scapes: from theory to application in breeding maize for drought tolerance. J Exp Bot
62:855868
Mueller D, Leyser O (2011) Auxin, cytokinin and control of shoot branching. Ann Bot
107:12031212
Nieminen K, Immanen J, Laxell M, Kauppinen L, Tarkowski P, Dolezal K, Tahtiharju S, Elo
A, Decourteix M, Ljung K, Bhalerao R, Keinonen K, Albert VA, Helariutta Y (2008)
Cytokinin signaling regulates cambial development in poplar. Proc Natl Acad Sci USA
105:2003220037
Olsen JE (2010) Light and temperature sensing and signalling in induction of bud dormancy in
woody plants. Plant Mol Biol 73:3747
Pin PA, Nilsson O (2012) The multifaceted roles of FLOWERING LOCUS T in plant develop-
ment. Plant, Cell Environ 35:17421755
Reymond M, Muller B, Leonardi A, Charcosset A, Tardieu F (2003) Combining quantitative
trait loci analysis and an ecophysiological model to analyze the genetic variability of the
responses of maize leaf growth to temperature and water deficit. Plant Physiol 131:664675
Richter GM, Acutis M, Trevisiol P, Latiri K, Confalonieri R (2010a) Sensitivity analysis for a
complex crop model applied to durum wheat in the Mediterranean. Eur J Agron 32:141150
Richter GM, Dailey GA, Riche AB (2010b) Identifying key process parameters in a sink-source
interaction model for energy crops. In: Wery J S-TI, Perrin A ed. Proceeding of the XI
Congress of the Eur Soc of Agronomy (Agro 2010). Montpellier, p 9798.
Rohde A, Prinsen E, De Rycke R, Engler G, Van Montagu M, Boerjan W (2002) PtABI3
impinges on the growth and differentiation of embryonic leaves during bud set in poplar.
Plant Cell 14:18851901
Rnnberg-Wastljung AC, Glynn C, Weih M (2005) QTL analyses of drought tolerance and
growth for a Salix dasyclados x Salix viminalis hybrid in contrasting water regimes. Theor
Appl Genet 110:537549
Rnnberg-Wastljung AC, Ahman I, Glynn C, Widenfalk O (2006) Quantitative trait loci for
resistance to herbivores in willow: field experiments with varying soils and climates. Entom
Exp Appl 118:163174
Ruttink T, Arend M, Morreel K, Storme V, Rombauts S, Fromm J, Bhalerao RP, Boerjan W,
Rohde A (2007) A molecular timetable for apical bud formation and dormancy induction in
poplar. Plant Cell 19:23702390
Rytter RM (1999) Fine-root production and turnover in a willow plantation estimated by differ-
ent calculation methods. Scand J Res 14:526537
Rytter RM (2001) Biomass production and allocation, including fine-root turnover, and annual N
uptake in lysimeter-grown basket willows. For Ecol Man 140:177192
Rytter RM, Hansson AC (1996) Seasonal amount, growth and depth distribution of fine roots in
an irrigated and fertilized Salix viminalis L-plantation. Biomass Bioenergy 11:129137
Rytter RM, Rytter L (1998) Growth, decay, and turnover rates of fine roots of basket willows.
Can J For Res-Rev Can Rech For 28:893902
Sanchez-Rodriguez C, Bauer S, Hematy K, Saxe F, Ibanez AB, Vodermaier V, Konlechner C,
Sampathkumar A, Rueggeberg M, Aichinger E, Neumetzler L, Burgert I, Somerville C,
Hauser M-T, Persson S (2012) CHITINASE-LIKE1/POM-POM1 and its homolog CTL2 are
glucan-interacting proteins important for cellulose biosynthesis in Arabidopsis. Plant Cell
24:589607
Sarkar P, Bosneaga E, Auer M (2009) Plant cell walls throughout evolution: towards a molecular
understanding of their design principles. J Exp Bot 60:36153635
130 A. Karp et al.
Sawa M, Kay SA (2011) GIGANTEA directly activates Flowering Locus T in Arabidopsis thali-
ana. Proc Natl Acad Sci USA 108:1169811703
Sawa M, Kay SA (2012) Circadian clocks and photoperiodism. Fukuoka igaku zasshi (=Hukuoka
acta medica) 103:2934
Schapendonk AHCM, Stol W, van Kraalingen DWG, Bouman BAM (1998) LINGRA, a sink/
source model to simulate grassland productivity in Europe. Eur J Agron 9:87100
Schrader J, Nilsson J, Mellerowicz E, Berglund A, Nilsson P, Hertzberg M, Sandberg G (2004)
A high-resolution transcript profile across the wood-forming meristem of poplar identifies
potential regulators of cambial stem cell identity. Plant Cell 16:22782292
Schuetz M, Smith R, Ellis B (2013) Xylem tissue specification, patterning, and differentiation
mechanisms. J Exp Bot 64:1131
Sennerby-Forsse L (1986) Seasonal variation in the ultrastructure of the cambium in young stems
of willow (Salix viminalis) in relation to phenology. Physiol Plant 67:529537
Sennerby-Forsse L (1995) Growth processes. Biomass Bioenergy 9:3543
Sennerby-Forsse L, Zsuffa L (1995) Bud structure and resprouting in coppiced stools of
Salix viminalis L, S. eriocephala Michx, and S. amygdaloides Anders. Trees-Struct Func
9:224234
Sennerby-Forsse L, Berggren B, Brunkener L, Fjell I (1984) Growth behaviour and anatomy of
meristems in Salix. In: Perttu K (ed) Ecology and management of forest biomass. Swed Univ
Agric Sci Rep 15:481501
Siriwardana NS, Lamb RS (2012) The poetry of reproduction: the role of LEAFY in Arabidopsis
thaliana flower formation. Int J Devel Biol 56:207221
Srikanth A, Schmid M (2011) Regulation of flowering time: all roads lead to Rome. Cell Mol
Life Sci 68:20132037
Stott KG (1961) The rooting pattern of the basket willowSalix triandra var. Black maul
at Long Asthon. In: annual report for long Ashton research station for 1961. Long Ashton
Research Station, Long Ashton, pp 187190
Triana F, Ragaglini G, Bonari E, Cerasuolo M, Richter GM (2011) Modelling the water balance
of different grass species used for bioenergy. Biomass Energy Crops IV 112:163170
Tsarouhas V, Gullberg U, Lagercrantz U (2002) An AFLP and RFLP linkage map and quantita-
tive trait locus (QTL) analysis of growth traits in Salix. Theor Appl Genet 105:277288
Tsarouhas V, Gullberg U, Lagercrantz U (2003) Mapping of quantitative trait loci controlling
timing of bud flush in Salix. Hereditas 138:172178
Tsarouhas V, Gullberg U, Lagercrantz U (2004) Mapping of quantitative trait loci (QTLs) affect-
ing autumn freezing resistance and phenology in Salix. Theor Appl Genet 108:13351342
Tuskan GA (2006) The genome of black cottonwood. Populus tricharpa Science 313:15961604
Verwijst T (1996a) Cyclic and progressive changes in short-rotation willow coppice systems.
Biomass Bioenergy 11:161165
Verwijst T (1996b) Stool mortality and development of a competitive hierarchy in a Salix vimi-
nalis coppice system. Biomass Bioenergy 10:245250
Vos J, Evers JB, Buck-Sorlin GH, Andrieu B, Chelle M, de Visser PHB (2010) Functional-
structural plant modelling: a new versatile tool in crop science. J Exp Bot 61:21012115
Ward SP, Salmon J, Hanley SJ, Karp A, Leyser O (2013) Using Arabidopsis to study shoot
branching in biomass willow. Plant Physiol 162:800811
Weih M (2009) Genetic and environmental variation in spring and autumn phenology of bio-
mass willows (Salix spp.): effects on shoot growth and nitrogen economy. Tree Physiol
29:14791490
Weih M, Rnnberg-Wastljung AC (2007) Shoot biomass growth is related to the vertical leaf
nitrogen gradient in Salix canopies. Tree Physiol 27:15511559
Weih M, Rnnberg-Wastljung AC, Glynn C (2006) Genetic basis of phenotypic correlations
among growth traits in hybrid willow (Salix dasyclados x S.viminalis) grown under two
water regimes. New Phytol 170:467477
Yang XH, Kalluri UC, DiFazio SP, Wullschleger SD, Tschaplinski TJ, Cheng ZMM, Tuskan GA
(2009) Poplar genomics: state of the science. Crit Rev Plant Sci 28:285308
Chapter 8
Camelina: An Emerging Oilseed Platform
for Advanced Biofuels and Bio-Based
Materials
Umidjon Iskandarov, Hae Jin Kim and Edgar B. Cahoon
Abstract Camelina (Camelina sativa (L.) Crantz) is a Brassicaceae oilseed crop

with valuable agronomic and biotechnological attributes that make it an attractive
renewable feedstock for biofuels and bio-based materials. Camelina seeds contain
3040% oil and can achieve oil yields per hectare that surpass established oilseed
crops such as soybean. Camelina is also productive under conditions of limited
rainfall and low soil fertility. As a short season, frost tolerant oilseed, Camelina
is amenable to double cropping systems and fallow year production. Simple, non-
labor intensive Agrobacterium-based transformation methods have recently been
described for Camelina that can be used in combination with breeding to rapidly
improve seed quality and agronomic traits to advance Camelina as a production
platform for biofuels and industrial feedstocks in geographical regions such as the
North American Great Plains that currently have little oilseed production for edi-
ble vegetable oils.
Keywords Camelina Oilseed crop Biofuels Bio-based materials Biodiesel

Fatty acids
U. Iskandarov(*) H. J. Kim(*) E. B. Cahoon

Center for Plant Science Innovation and Department of Biochemistry, University of
Nebraska-Lincoln, E318 Beadle Center, 1901 Vine Street, Lincoln, NE 68588, USA
e-mail: uiskandarov2@unl.edu
H. J. Kim
e-mail: haejin.kim@unl.edu
E. B. Cahoon
e-mail: ecahoon2@unl.edu
132 U. Iskandarov et al.
8.1Introduction
Camelina sativa or Camelina, known also as false flax or gold of pleasure, is an

annual oilseed species of the Brassicaceae family. Camelina has received growing
interest as biofuel crop for the production of vegetable oils for biodiesel and avia-
tion fuel because of its productivity in geographic regions that are not currently
used for large-scale oilseed production and the possibility of growing Camelina in
farming systems (e.g., double cropping) that do not compete with crops for food
production. Camelina currently has a number of seed quality and agronomic issues
that limit its wider use for biofuel and possibly other industrial feedstock produc-
tion. These limitations can now be addressed by advances in biotechnology and
the increasing availability of genomic resources to facilitate breeding.
Camelina is native to Eastern Europe and Central Asia (Putnam et al. 1993).
It has been cultivated in Europe since the Bronze age, as early as 1500400 BC
(Bouby 1998; Zubr 2010). It was introduced to North America from Europe most
likely as a weed along with flax and is well adapted to Southern Canada, Northern
Great Plains and Pacific Northwest of the U.S. Camelina grows 0.31m high, stems
are smooth and branched with arrow shaped leaves that are 58cm long (Fig.8.1).
Flowers are small and consist of four pale or greenish yellow petals and give rise
to pear-shaped pods (510mm) containing as many as 16 seeds. Seeds are small
Fig.8.1Camelina sativa
(Camelina). a Flowering
Camelina plant. b Mature
pods. c Comparison of
Arabidopsis (Arabidopsis
thaliana), Camelina, and
rapeseed (Brassica napus)
seed sizes
8 Camelina: An Emerging Oilseed Platform 133
yellowbrown or brown, oblong, rough with rigid surface (Putnam et al. 1993).
Seeds contain 3040% oil on dry weight basis of which 64% is polyunsaturated,
including 3045% of the omega-3 fatty acid -linolenic acid (Vollmann et al. 1996).
Historically Camelina has been valued for the vegetable oil extracted from
its seeds (Moloney et al. 1998; Jaskiewicz and Matyka 2003; Flachowsky etal.
2011). The ancient Romans used Camelina oil as massage oil, lamp fuel, as
well as cooking oil. In modern times, Camelina oil has gained a niche market,
particularly in Europe, for its nutraceutical value because of its high content of
-linolenic acid (Hurtaud and Peyraud 2007; Ehrensing and Guy 2008). Camelina
meal obtained after oil extraction, contains 1011% fiber, 1014% oil, and about
40% protein making it a valuable product for use (Korsrud et al. 1978; Onyilagha
et al. 2012) in cattle, chicken and hog feed (Ehrensing and Guy 2008). However,
due to the presence of glucosinolates (1923mol/g) in Camelina meal regula-
tions require limited daily use to avoid negative impacts on livestock productiv-
ity although the amount is not higher than that of the widely used canola meal
(Matthus and Angelini 2005; Moser 2010). U.S. Food and Drug Administration
regulations currently restrict the use of Camelina meal to10% of the total diet
of beef cattle, broiler chickens, and laying hens and2% of the diet of growing
swine for United States livestock production.
8.2Agronomic Properties
Camelina has recently gained renewed interest as a low input biofuel feedstock because
of its minimal agronomic input requirements for productivity on marginal lands with
limited fertility and water resources. Camelina requires low to moderate amounts of
fertilizers and is productive with nitrogen and phosphorus levels of as low as 90100
and 67kg/ha, respectively (McVay and Lamb 2007; Ehrensing and Guy 2008; Sipalova
et al. 2011; Solis et al. 2013). Field trials with cambic chernozem soil in Timisoara,
Romania showed that only addition of the fertilizers nitrogen (100kg/ha) and phos-
phorus (60kg/ha) increased seed yield from 932 to 1,813kg/ha and oil up to 584kg/ha
increasing the oil yield by 25% when plants were sowed at 25cm row distance which
was found to be optimal (Imbrea et al. 2011). Similar doses of nitrogen fertilizer were
recommended for Camelina growth to achieve optimal yield based on the results of the
laboratory experiments (Sipalova et al. 2011).
Despite being a cool season crop, Camelina is well suited for dryland crop-
ping systems in which soil moisture and rainfall can be maximized by planting in
early months of spring depending on location. For example, as high as 1,912kg/ha
of grain and 506kg/ha oil yield were achieved in dryland conditions near Havre,
Montana USA, while grain yield of 3,250kg/ha was obtained in Austria if water
was not limiting throughout vegetation period (Vollmann et al. 1996; McVay
and Lamb 2007; Ehrensing and Guy 2008). The 2year trials in arid zones of
Southwestern USA showed that Camelina can be a low water use crop. Camelina
grown from January to May in Maricopa, Arizona yielded in over 1,500kg/ha seed
with maximum seasonal water use of just 4749cm, furthermore the seed yield
loss was not significant unless the soil water depletion before irrigation reached
70% and higher (Hunsaker et al. 2011).
Camelina has traditionally been grown without pesticides and has allelo-
pathic effects on weed species (Ehrensing and Guy 2008; Gesch and Cermak
2011). Herbicide research on Camelina is currently ongoing. Being a minor
weed, Camelina is usually not a problem in other crops and does not have seed
dormancy. Research showed that by planting Camelina seeds in winter or early
spring, herbicide use can be avoided since Camelina seeds can germinate at
low temperatures and seedlings are frost tolerant (Robinson 1987) and suppress
many weed species, except perennial weeds, especially if seeded at high density
(Ehrensing and Guy 2008; Gesch and Cermak 2011). Camelina was shown to be
a viable winter crop for the Northern Corn Belt of the U.S., where seed yields as
high as 1,317kg/ha and oil yields as high as 420g/kg were obtained from fall
sown crop promoting good weed suppression (Gesch and Cermak 2011). Thus
sowing does not require pre-emergence weed control, greatly reducing both pro-
duction costs and environmental damage. Susceptibility of Camelina to herbicides
inhibiting acetolactate synthase (ALS), that are commonly used for wheat cultiva-
tion in the Pacific Northwest of the U.S., has limited cultivation of Camelina as an
oilseed crop in the area (Hanson et al. 2004; Pavlista et al. 2011).
Unlike other Brassicaceae crops such as rapeseed, Camelina does not require
insecticide application. Camelina is resistant to crucifer insect pests due to large
concentrations of the insect deterrent quercetin glycosides in its tissues (Onyilagha
et al. 2012; Naranjo and Stefanek 2012). Camelina is also highly resistant to
blackleg disease caused by the fungus Leptosphaeria maculans, which is a major
pathognen of many Brassicaceae crops, such as canola. Camelina is susceptible
to downy Peronospora Camelinae (downy mildew), Alternaria brassicae and the
saprotrophic fungus Rhizoctonia, but none of these pathogens has been reported
to cause major yield losses in Camelina (Robinson 1987; Salisbury et al. 1995;
Ehrensing and Guy 2008).
Because it has a relatively short growth cycle (80100days) and frost tolerance,
Camelina can be used in double cropping systems during cool periods of the year
(Gesch and Archer 2012). Field trials showed that short season cultivars of soy-
bean could be double cropped after winter Camelina in the upper U.S. Midwest
(Gesch and Archer 2012). The net return from the Camelina-soybean double crop
was higher than that of mono-cropped soybean in the period analyzed (Gesch and
Archer 2012). Camelina can also be grown following wheat, barley, peas and len-
tils, but should not be planted following Brassicaceae crops to avoid increased
risks by pests and diseases (Fleenor 2011). In addition, Camelina has been con-
sidered as a tertiary crop for rotations in northeastern Colorado where wheat is
rotated with crops such as corn (Brandess 2012). In this region, Camelina could
be planted in October after the first crop in year one and harvested before July
of year two. This could then be followed by land recuperation period of July to
mid-September when winter wheat could be planted and harvested in July of year
three. This is followed by a fallow period until April or May of the year four when
the rotation restarts (Brandess 2012). This rotation would allow farmers to grow
three crops versus two in three years (Brandess 2012). Since it is best adapted to
cooler climates, Camelina could also be grown in winter in areas with mild win-
ters (Gesch and Cermak 2011).
Although the oil content of seeds is lower than that of Brassica napus on dry
weight basis, the oil yield per hectare can reach as high as that of B. napus if good
agronomic management is practiced (Putnam et al. 1993; Imbrea et al. 2011).
Camelina has a low seeding rate (as low as 35kg/ha) compared to other agro-
nomic crops, including canola to establish dense stands (McVay and Lamb 2007;
Pilgeram et al. 2007). In addition, existing equipment that is used for harvesting and
processing of other crops can be adapted to Camelina (Brandess 2012). These agri-
cultural attributes of Camelina give it compelling properties and make it a favorable
oilseed crop to be grown in agronomically demanding lands with low-inputs.
8.3Genetic Improvement: Variety Selection, Breeding,

Genomic Resources, Biotechnology
Limiting the full potential of Camelina as a biofuel oilseed crop is the need to
improve a number of agronomic, yield, and oil quality traits. In contrast to other
Brassicaceae oilseeds such as canola and rapeseed, Camelina has not undergone
extensive breeding and only a relatively small number of cultivars are available
for commercial production. From an agronomic production standpoint, improve-
ment in traits such as heat tolerance, downy mildew resistance, water and nitrogen
use efficiencies, and herbicide resistance are desirable. For biodiesel and industrial
uses such as bio-based lubricants, enhancement in seed oil content from the cur-
rent 3040% of seed weight to levels of 4050% of seed weight, as is currently
found in elite rapeseed germplasm, is a major target for Camelina crop improve-
ment. An additional target for biodiesel and bio-lubricants is the reduction of the
high polyunsaturated fatty acid content of the seed oil [3539% linolenic acid
(18:3) and 2025% linoleic acid (18:2)] and replacement with high content of the
more oxidatively stable monounsaturated fatty acid oleic acid (18:1). Furthermore,
reductions in seed glucosinolate levels would allow for increased use of Camelina
meal for livestock production.
Varietal selection and screening of germplasm following mutagenesis are
among the approaches used to date for Camelina crop improvement. Several
Camelina varieties have been selected for higher oil content and improved fatty
acid composition. For example, the cultivar Blaine Creek is richer in -3 fatty
acids, while Suneson has 23% higher oil content, and is rich in -linolenic
acid (Ehrensing and Guy 2008). Lines with resistance to acetolactate synthase
(ALS)-targeting herbicides imazethapyr and sulfosulfuron and altered seed fatty
acid composition, including increased oleic acid content, have also been identi-
fied screening of mutagenized populations (Vollamnn et al. 1997; Buchsenschutz-
Nothdurft et al. 1998; Kang et al. 2011; Walsh et al. 2012).
Recent advances in Camelina genomics are also providing avenues for Camelina
improvement through marker-assisted breeding. Molecular genetic maps have been
assembled for Camelina using random amplified polymorphic DNA (RAPD) markers
and amplified fragment length polymorphisms (Vollmann et al. 2005; Gehringer etal.
2006). These maps have been used to localize QTLs for agronomic characteristics
such as seed yield, oil content, 1,000-seed weight, and plant height (Gehringer et al.
2006). More recent AFLP fingerprinting data using 53 accessions from different ori-
gins showed a high genetic diversity in the species which could offer opportunities for
breeding (Ghamkhar et al. 2010). The chromosome number of Camelina is 2n=40,
which was confirmed by linkage map using 157 AFLP markers and 3 Brassica SSR
markers (Gehringer et al. 2006). Genetic mapping based on AFLP, SSR and ILP mak-
ers indicated that Camelina is a hexaploid (Hutcheon et al. 2010). This was supported
by isolation of three copies of FATTY ACID DESATURASE 2 (FAD2) and FATTY
ACID ELONGASE 1 (FAE1) genes, both of which are single copy in Arabidopsis
(Gehringer et al. 2006; Galasso et al. 2011). Like other important crops, polyploidy
of Camelina will likely complicate efforts to develop molecular markers and assemble
whole genome sequence. Advanced technologies for molecular genetics and genom-
ics, including RNA-seq and next-generation genome and transcriptome sequencing,
will likely provide unprecedented opportunities to accelerate improvement of agro-
nomic and seed quality traits for Camelina (Varshney et al. 2009; Edwards et al. 2012).
Complementing the impact of breeding on crop improvement, Camelina
is highly amenable to biotechnological enhancement through the use of
Agrobacterium tumefaciens-mediated transformation. Camelina can be easily trans-
formed using protocols similar those routinely used for Arabidopsis thaliana trans-
formation. These methods include vacuum infiltration of flowers with a solution of
Agrobacterium harboring a binary vector that contains the desired transgene (Lu
and Kang 2008), or by simple floral dip with the Agrobacterium solution (Liu et al.
2012). Plants with Agrobacterium-infiltrated or -dipped flowers are grown to matu-
rity. Seeds obtained from these plants are then screened to identify those containing
the transgene. For this process, genes for resistance to antibiotics (e.g., kamamycin
and hygromycin) or herbicides (e.g., glufosinate) can be used as selection markers
for obtaining transgenic plants by screening of seeds on media containing the selec-
tive agent or by spraying seedlings with the selective agent in the case of herbicides
(Lu and Kang 2008; Liu et al. 2012). Fluorescent protein selection markers such as
DsRed under control of a seed-specific or constitutive promoter can also be used to
identify transgenic seeds based on fluorescence of seeds with equipment as low tech
as a green LED flashlight and red camera filter (Lu and Kang 2008). A wide variety
of seed specific promoter/3UTR cassettes can easily be inserted into binary vectors
to express several candidate genes to modify seed oil traits. With these transforma-
tion methods and metabolic engineering toolbox, transgenic seeds can be obtained
in as little as 68 weeks following Agrobacterium infiltration or dipping of flow-
ers. Unlike transformation protocols for most crops, Camelina transformation can
be done with minimal labor input and without the need for specialized technical
skills. As such, Agrobacterium-based transformation offers a relatively simple and
rapid, cost-effective approach for improvement of agronomic and seed quality traits.
Recent biotechnological efforts to improve the agronomic properties of Camelina

have included transgenic expression of Arabidopsis purple acid phosphatase 2 that
resulted in increased seed size and yield and faster growing plants relative to non-
transformants. In addition, a recent report described the enhancement of the oleic
acid content, a desirable biofuel trait, in Camelina seeds by anti-sense suppression
using an inverted portion of the Camelina CsFAD2-1 gene under the control of the
seed-specific promoter for the phaseolin gene (Kang et al. 2011). The resulting
transgenic seeds contained 3851% oleic acid compared to 1318% oleic acid in
seeds from non-transformed plants (Kang et al. 2011). In this study, seeds from a
mutant of the CsFAD2-2 locus obtained from random mutagenesis that contained
a premature stop at the Trp288 codon had~27% oleic acid (Kang et al. 2011). An
interpretation of this result is that two or more of the three known FAD2 loci in
Camelina contribute to the desaturation of oleic acid in seeds. Because of the high
identity of these genes, it is possible to use antisense or RNA interference suppres-
sion with sequence from only one of these genes to get a mid- to high-oleic acid
trait in Camelina seeds. It can be envisioned that additional enhancements in oleic
acid content can be achieved through transgenic suppression of the FAE1 genes
that are responsible for the elongation of oleic acid (18:1) to gondoic acid (20:1).
Moreover, genes from other species could be transferred to Camelina to obtain
additional biofuel-type traits, such as short- and medium-chain fatty acid-specific
FatB-type thioesterases to achieve an oil functionality mimicking Jet A1 fuel. For
industrial uses of Camelina oil, the castor bean (Ricinus communis) fatty acid
hydroxylase has been successfully transferred to Camelina to produce ricinoleic
acid and other hydroxy fatty acid in the seed oil of transformants (Lu and Kang
2008). As Camelina crop improvement progresses, it is likely that these efforts may
involve a combination of varietal selection, mutagenic breeding, and biotechnologi-
cal approaches. Indeed, the simple transformation protocols for Camelina hold con-
siderable promise for rapid genetic improvement of this crop.
8.4Current and Future Prospects for Commercial

Production
Despite its considerable potential as a low input oilseed for biofuel production,
Camelina has yet to see extensive commercial production. In the United States, large-
scale production of Camelina has largely been restricted to the state of Montana
where 8,400 ha (20,800 acres) were grown in 2009 (Anonymous 2011a). Spurring
the recent interest in Camelina has been successful tests that have used Camelina oil
as an ingredient of aviation fuel for commercial airliners and military jets. To stimu-
late increased commercial planting of Camelina, the U.S. Department of Agriculture
Farm Service Agency announced in July 2011 a program targeted for the states of
California, Washington, and Montana under the Biomass Crop Assistance Program
to provide 5year contracts for the production of up to 20,200 ha (50,000 acres)
of Camelina for aviation fuel or other biomass conversion (Anonymous 2011b).
Table8.1Fatty acid Oil source Fatty acid composition (% of total fatty acids)
composition of seed oils
16:0 18:0 18:1 18:2 18:3 20:0 20:1 22:1
of Camelina, soybean,
sunflower, and rapeseed Camelina 7.5 4.0 14.9 21.4 31.7 5.1 11.8 3.3
Soybean 11.6 4.6 23.9 52.0 6.8 0.3 0.2 0.4
Sunflower 6.1 6.4 24.8 61.7 0.1 0.4 0.2 0.3
Rapeseed 3.3 1.7 18.1 14.1 7.7 2.8 8.2 44.2
More widespread production of Camelina in the North American Great Plains and
the Pacific Northwest of the United States will undoubtedly require ready markets
for Camelina oil and meal with cost-competitive pricing as well as the development
of more extensive infrastructure for crushing of Camelina seeds and conversion of
Camelina oil to biodiesel, aviation, or other fuel (Table8.1).
Acknowledgments Research in the Cahoon lab for Camelina genetic improvement is
supported by grants from the Center for Advanced Biofuel Systems (CABS), an Energy Frontier
Research Center funded by the U.S. Department of Energy, Office of Science, Office of Basic
Energy Sciences under Award Number DE-SC0001295, U.S. Department of Agriculture
Agriculture and Food Research Initiative 2009-05988, and NSF Plant Genome IOS 0701919.
References
Anonymous (2011a) 2010 Camelina Crop United States Department of Agriculture

National Agricultural Statistics Service, Washington, D.C. http://www.nass.usda.gov/
Statistics_by_State/Montana/Publications/Press_Releases_Crops/camelina.pdf
Anonymous (2011b) Fact sheet: biomass crop assistance programproject area number 8
Camelina growers in California. Montana and Washington United States Department
of Agriculture Farm Service Agency, Washington, D.C. http://www.fsa.usda.gov/
Internet/FSA_File/bcap_8_fact_sheet.pdf
Bouby L (1998) Two early finds of gold-of-pleasure (Camelina sp.) in middle Neolithic and
Chalcolithic sites in western France. Antiquity 72:391398
Brandess A (2012) Modeling the profitability of Camelina sativa as a biofuel feedstock in eastern
Colorado. Colorado State University
Buchsenschutz-Nothdurft A, Schuster A, Friedt W (1998) Breeding for modified fatty acid compo-
sition via experimental mutagenesis in Camelina sativa (L.) Crtz. Ind Crops Prod 7:291295
Edwards D, Batley J, Snowdon RJ (2012) Accessing complex crop genomes with next-genera-
tion sequencing. Theor Appl Genet (in press)
Ehrensing DT, Guy SO (2008) Camelina. Oregon State University, Corvalis. http://
extension.oregonstate.edu/catalog/pdf/em/em8953-e.pdf
Flachowsky G, Langbein T, Bhme H, Schneider A, Aulrich K (2011) Effect of false flax expel-
ler combined with short-term vitamin E supplementation in pig feeding on the fatty acid pat-
tern, vitamin E concentration and oxidative stability of various tissues. J Anim Physiol Anim
Nutr 78:187195
Fleenor RA (2011) Plant guide for Camelina (Camelina sativa) USDA-natural resources conser-
vation service, Spokane. http://plants.usda.gov/plantguide/pdf/pg_casa2.pdf
Galasso I, Manca A, Braglia L, Martinelli T, Morello L, Breviario D (2011) h-TBP: an approach
based on intron-length polymorphism for the rapid isolation and characterization of the multiple
members of the -tubulin gene family in Camelina sativa (L.) Crantz. Mol Breed 28:635645
Gehringer A, Friedt W, Luhs W, Snowdon RJ (2006) Genetic mapping of agronomic traits in

false flax (Camelina sativa subsp. sativa). Genome 49:15551563
Gesch RW, Cermak SC (2011) Sowing date and tillage effects on fall-seeded Camelina in the
Northern Corn Belt. Agron J 103:980987
Gesch RW, Archer DW (2012) Double-cropping with winter Camelina in the northern Corn Belt
to produce fuel and food. Ind Crops Prod (in press)
Ghamkhar K, Croser J, Aryamanesh N, Campbell M, Konkova N, Francis C (2010) Camelina
(Camelina sativa (L.) Crantz) as an alternative oilseed: molecular and ecogeographic analy-
ses. Genome 53:558567
Hanson B, Park K, Mallory-Smith C, Thill D (2004) Resistance of Camelina microcarpa to ace-
tolactate synthase inhibiting herbicides. Weed Res 44:187194
Hunsaker D, French A, Clarke T, El-Shikha D (2011) Water use, crop coefficients, and irrigation
management criteria for Camelina production in arid regions. Irrig Sci 29:2743
Hutcheon C, Ditt RF, Beilstein M, Comai L, Schroeder J, Goldstein E, Shewmaker CK, Nguyen
T, De Rocher J, Kiser J (2010) Polyploid genome of Camelina sativa revealed by isolation of
fatty acid synthesis genes. BMC Plant Biol 10:233
Hurtaud C, Peyraud J (2007) Effects of feeding Camelina (seeds or meal) on milk fatty acid
composition and butter spreadability. J Dairy Sci 90:51345145
Imbrea F, Jurcoane S, Halmajan H, Duda M, Botos L (2011) Camelina sativa: a new source of
vegetal oils. Rom Biotechnol Lett 16
Jaskiewicz T, Matyka S (2003) Application of Camelina sativa, its seeds, extrudate and oil cake
in diets for broiler chickens and the effect on rearing indices and carcass quality. Ann Anim
Sci Suppl 2:181184
Kang J, Snapp AR, Lu C (2011) Identification of three genes encoding microsomal oleate desatu-
rases (FAD2) from the oilseed crop Camelina sativa. Plant Physiol Biochem 49:223229
Korsrud GO, Keith MO, Bell JM (1978) Comparison of nutritional-value of crambe and
Camelina seed meals with egg and casein. Can J Anim Sci 58:493499
Liu X, Brost J, Hutcheon C, Guilfoil R, Wilson AK, Leung S, Shewmaker CK, Rooke S, Nguyen
T, Kiser J, De Rocher J (2012) Transformation of the oilseed crop Camelina sativa by
Agrobacterium-mediated floral dip and simple large-scale screening of transformants. In
Vitro Cell Devel Biol-Plant 48:462468
Lu C, Kang J (2008) Generation of transgenic plants of a potential oilseed crop Camelina sativa
by Agrobacterium-mediated transformation. Plant Cell Rep 27:273278
Matthus B, Angelini LG (2005) Anti-nutritive constituents in oilseed crops from Italy. Ind Crops
Prod 21:8999
McVay K, Lamb P (2007) Camelina production in Montana. Montana State University Extension,
Bozeman. http://msuextension.org/publications/AgandNaturalResources/MT200701AG.pdf
Moloney A, Woods V, Crowley J (1998) A note on the nutritive value of Camelina meal for beef
cattle. Ir J Agric Food Res 243247
Moser BR (2010) Camelina (Camelina sativa L.) oil as a biofuels feedstock: golden opportunity
or false hope? Lipid Technol 22:270273
Naranjo SE, Stefanek MA (2012) Feeding behavior of a potential insect pest, Lygus hesperus, on
four new industrial crops for the arid southwestern USA. Ind Crops Prod 37:358361
Onyilagha JC, Gruber MY, Hallett RH, Holowachuk J, Buckner A, Soroka JJ (2012) Constitutive
flavonoids deter flea beetle insect feeding in Camelina sativa L. Biochem Syst Ecol 42:128133
Pavlista A, Isbell T, Baltensperger D, Hergert G (2011) Planting date and development of spring-
seeded irrigated canola, brown mustard and Camelina. Ind Crops Prod 33:451456
Pilgeram AL, Sands DC, Boss D, Dale N, Wichman D, Lamb P, Lu C, Barrows R, Kirkpatrick
M, Thompson B (2007) Camelina sativa, a Montana omega-3 and fuel crop. Issues New
Crops New Uses 129131
Putnam D, Budin J, Field L, Breene W (1993) Camelina: a promising low-input oilseed. In:
Janick J, Simon J (eds) New crops. Wiley, New York, pp 314322
Robinson RG (1987) Camelina: a useful research crop and a potential oilseed crop. University of
Minnesota
Salisbury P, Ballinger D, Wratten N, Plummer K, Howlett B (1995) Blackleg disease on oilseed

Brassica in Australia: a review. Anim Prod Sci 35:665672
Sipalova M, Losak T, Hlusek J, Vollmann J, Hudec J, Filipcik R, Macek M, Kracmar S (2011)
Fatty acid composition of Camelina sativa as affected by combined nitrogen and sulphur fer-
tilisation. Afr J Agric Res 6:39193923
Solis A, Vidal I, Paulino L, Johnson BL, Berti MT (2013) Camelina seed yield response to nitro-
gen, sulfur, and phosphorus fertilizer in South Central Chile. Ind Crops Prod 44:132138
Varshney RK, Nayak SN, May GD, Jackson SA (2009) Next-generation sequencing technologies
and their implications for crop genetics and breeding. Trends Biotechnol 27:522530
Vollmann J, Damboeck A, Eckl A, Schrems H, Ruckenbauer P (1996) Improvement of Camelina
sativa, an underexploited oilseed, vol 1. ASHS Press, Alexandria
Vollmann J, Damboeck A, Baumgartner S, Ruckenbauer P (1997) Selection of induced mutants
with improved linolenic acid content in Camelina. Lipid/Fett 99:357361
Vollmann J, Grausgruber H, Stift G, Dryzhyruk V, Lelley T (2005) Genetic diversity in Camelina
germplasm as revealed by seed quality characteristics and RAPD polymorphism. Plant Breed
124:446453
Walsh DT, Babiker EM, Burke IC, Hulbert SH (2012) Camelina mutants resistant to acetolactate
synthase inhibitor herbicides. Mol Breed 30:10531063
Zubr J (2010) Carbohydrates, vitamins and minerals of Camelina sativa seed. Nutr Food Sci
40:523531
Chapter 9
Perspectives in Brazil of the Contribution
of Palm Trees to Biodiesel Production
Janaina M. Meyer and Antonio Salatino
Abstract The awareness of the depletion and contamination derived from fossil
fuels and the resultant environmental crisis has led to the recognition of the neces-
sity of research and development aiming the production of biofuels. An escalation
of oil prices has been the outcome of political instabilities in areas of the world
which traditionally have been massive petroleum providers, allied to oil deficits
and economic crises. The present paper discusses the potentialities of palm species
native to Brazil as feedstock of biodiesel.
9.1Introduction
The awareness of the depletion and contamination derived from fossil fuels and
the resultant environmental crisis has led to the recognition of the necessity of
research and development aiming the production of biofuels. An escalation of oil
prices has been the outcome of political instabilities in areas of the world which
traditionally have been massive petroleum providers, allied to oil deficits and eco-
nomic crises. This has contributed to a world claim for alternatives to fossil oils,
among which products derived from plants has been the aim of great expectation,
particularly in developing countries (Zhuang et al. 2011). Thus, new strategies for
power generation have become urgent all over the world. Some alternatives based
on plant products have been implemented, such as the production of ethanol from
corn and sugar-cane, as well as biodiesel from seed oils. Biofuels not only may
reduce the dependency on imported mineral oil, but also have the advantage of
giving off emissions with lower load of damaging environmental pollutants.
J. M. Meyer A. Salatino(*)
Department of Botany, Institute of Biosciences, University of So Paulo,
Rua do Mato 277, So Paulo, SP 05508-090, Brazil
e-mail: asalatin@ib.usp.br
142 J. M. Meyer and A. Salatino
However, Brazil and most developing countries face problems of inefficien-

cies regarding transport systems and energy distribution. Especially in countries
with continental dimensions such as Brazil, these constraints tend to give rise to
wide isolated areas, where many people remain with no possibilities to advance
toward biomass utilization and production of liquid fuels, gas and electricity
(REN21 2005). These difficulties and the success of the sugar-cane ethanol pro-
gram encouraged the Brazilian government to create in 2003 the National Program
for Production and Use of Biodiesel.
Biodiesel is a fuel prepared from vegetable oil or animal fat (triglycerides), by
means of a process of transesterification with alcohols, mainly methanol or eth-
anol, using a catalyst (sodium or potassium hydroxide). The product obtained is
a mixture of methyl or ethyl esters of fatty acids, plus glycerol as a by-product
(Buckeridge and Salatino 2010). Finding commercial uses for the latter substance
is important to achieve reductions in the final cost of biodiesel. Glycerol may
be used in chemistry and livestock industries. It has recently been proposed the
fermentative conversion of glycerol in ethanol by Escherichia coli (Yasdani and
Gonzales 2008). A study by the US Environmental Protection Agency (EPA 2002)
concluded that the addition of biodiesel to mineral diesel (ecodiesel) reduces pol-
lutant emission from the engine exhaust. This effect is due to the higher oxygen
content in biodiesel, which reduces the amount of unburned hydrocarbons, carbon
monoxide and particulate material in the engine exhausts. A reduction also occurs
in the amounts of sulfur and aromatic pollutants in the emissions. The amount of
nitrogen oxides (NOx) tend to be higher in ecodiesel emissions, but this problem
may be circumvented by tuning the engines properly (Sze et al. 2007). Xue et al.
(2011) reported similar amounts of NOx in biodiesel and mineral diesel emissions.
The production of biodiesel worldwide has grown considerably. It was neg-
ligible in 2000 and has grown incessantly since then, with a strong ramping up
beginning in 2005 (Fig.9.1). A combination of factors has accounted for the
increase in biodiesel production. First, the stead threat of sudden increases in min-
eral oil prices may be pointed out. Second, the independence from imported fos-
sil energy sources may bring about some economic security. Also important has
been the development of new technologies and introduction of alternative feed-
stocks for biodiesel production. A report from Pike Research (2011) estimates that
revenues derived from biodiesel production will be three times as much in 2020
(US$ 71billion) as they were in 2010 (US$ 18.4billion).
Given the environmental and economic perspectives, it is expected for the next
decades a continuous rise in the production of biodiesel in emerging countries,
mainly China and Brazil. Forecasts assume that Brazilian biodiesel production
will surpass the European production by 2015. By 2020, it is estimated that 10%
of all road diesel in BRICs countries (Brazil, Russia, India and China) will be bio-
diesel (Emerging Market Online 2012).
Brazilian government claims that the 2003 program for biodiesel production was
conceived with focus on regional development and social inclusion. Official data
assumes that thousands of jobs in familiar agriculture have recently been the result
of implementation of the program for cultivation of oleaginous plants for biodiesel
9 Perspectives in Brazil of the Contribution of Palm Trees 143
Fig.9.1Evolution of the world biodiesel production. Sources International energy agency,

Agncia Nacional de Petrleo (Brazil) and United States Department of Agriculture
production. Cultivation of plants as biodiesel feedstock in Brazilian regions most seri-

ously affected by economic and social problems (mainly in the north and northeast)
has contributed to bolster a modest but detectable lessening of regional disparities
inside the country. Companies buying raw material from familiar farmers for bio-
diesel production enjoy tax privileges and government financial support, including a
Social Fuel Stamp. They must, however, assume the commitment to purchase the total
production at pre-set prices, ensuring thus a financial safety for the low-scale farmer.
The production of biodiesel in Brazil has considerably risen in the last few
years. From a virtually null production up to 2006, the production rapidly grew
to an amount comparable to the output of France and Germany (world leaders) in
2010 (Fig.9.1). From a current mandatory blend of 3% (B3), expected to rise to
B5 in 2013, the hopeful expectation for 2015 is B20 (European Biofuels 2012).
Biodiesel cost optimization in Brazil requires that the distance between feed-
stock plantations and final consumers have to lie inside a 200km radius, in order
to reduce costs derived from transportation (Dias 2007). A means to help achiev-
ing this aim is the bioprospection of oleaginous potential sources as alternatives to
the species currently exploited as feedstock, most of them cultivated in huge mon-
ocultures considerably harmful to the environment. The extent of the Brazilian
territory and the wide diversity of its flora and habitats provide immense opportu-
nities to find alternative sources of biodiesel feedstock.
9.2Current Biodiesel Feedstock Sources in Brazil
Worldwide, biodiesel is produced mainly from vegetable oils (chiefly palm oil,
soybean and canola) and secondarily from animal fat and microalgae. In Brazil,
biodiesel is also produced chiefly from seed oils. Major sources are soybean
Fig.9.2Locations of
plantation of oleaginous
plants in Brazil for provision
of biodiesel feedstock.
Sources AGROPALMA,
ABIOVE and ABRAPA
(Glicyne max L., Leguminosae), cottonseed (Gossypium hirsutum L., Malvaceae)

and palm oil (Elaeis guineensis Jacq., Arecaceae). Minor contributors are sun-
flower oil (Helianthus annuus L., Asteraceae) and the fruit of the tucum palm tree
(Astrocaryum aculeatum Meyer, Arecaceae). Except for the latter source, the four
other species are not native to Brazil, despite the megadiversity of the country (the
widest in the world).
Soybean is a major biodiesel feedstock in many countries and Brazil is its
second largest producer. According to the Brazilian Institute of Geography and
Statistics, the cultivated area of soy plants in Brazil reaches up to 24.5millionha
(IBGE 2012), or 2,9% of the territory. The regions mostly used for soybean cul-
tivation are the south, southeast and central-west (Fig.9.2). The contribution to
biodiesel production of soybean oil in Brazil is 7080%. The second important
feedstock in Brazil is cottonseed oil, although contributing far below soybean oil:
35% (ANP 2012). Cottonseed plants are cultivated in distinct parts of the coun-
try, such as southeast, northeast and central-west (Fig.9.2).
The dependence on one or two major biodiesel feedstock is neither reliable nor
feasible, especially taking into consideration the expectation of the Brazilian gov-
ernment of a continuous and rapid increase of biodiesel blending (B20 in 2020).
For this reason, incentives have been invested in cultivation of several other spe-
cies, according to traditional and climatic characteristics of the Brazilian regions.
By encouraging the production of feedstock at many points of all regions of the
country, the program aims also to optimize costs of its transportation to biodiesel
plants. Palm oil tree is a species native from Africa. It has been one of the major
crops in tropical countries in the New and Old Worlds for production of edible
oil. The species has been assumed in Brazil as an important alternative for pro-
duction of oil in regions not suitable for cultivation of soybean and where it has
been cultivated and consumed traditionally: the northern and northeastern regions
(Fig.9.2). The tucum palm tree (Astrocaryum aculeatum, Arecaceae) is native in
the Amazon. It has been cultivated in several localities of the Brazilian Northern
regions as biodiesel feedstock (Fig.9.2), but achieving so far very little contribu-
tion to the national production. Jatropha curcas oil (Euphorbiaceae) has been rated
as one of the most promising feedstock for biodiesel production in the tropical
world. It was assumed as highly productive, drought resistant and requiring lit-
tle care for cultivation. Official funds were invested in several tropical countries
for cultivation of jatropha plants aiming the biodiesel production. For several rea-
sons (including ecological and economic unexpected problems) the former opti-
mistic forecasts were frustrated (Axelsson and Franzn 2010). Incentives from the
Brazilian government in the last decade has encouraged programs for cultivation
of jatropha plants in several areas of Brazil (Fig.9.2), and also the set-up of a bio-
diesel plant exclusively for production of jatropha biodiesel. However, the contri-
bution in Brazil from jatropha oil for biodiesel production has been negligible.
Given the so far limited resources being exploited, mostly from non-native spe-
cies, in addition to the immense potential possibilities offered by the diversity of
the Brazilian flora, bioprospection of alternative sources of biodiesel feedstocks
is crucial for technological development and improvement of the social-economic
condition of people from Brazilian less-favored areas.
9.3Potentialities of Oils from Palm Trees as Biodiesel

Feedstock
The family of the palm trees (Arecaceae) is one of the largest among the mono-
cotyledons, comprising about 1,500 species (Henderson et al. 1995). Palm trees
are one of the main physiognomic characteristics of tropical forests from both New
and Old Worlds. There are approximately 200 species of palms in the Brazilian
flora (Souza and Lorenzi 2005). Palm species are distributed in all Brazilian eco-
systems. The Amazon and Atlantic forests are among the world ecosystems with
wider diversity of palm species (Cintra et al. 2005). Examples of palm trees from
Brazilian rain forests are Euterpe oleracea (the aa palm) and Astrocaryum sci-
ophilum. Euterpe edulis, Mauritia flexuosa, Orbygnia phalerata and Syagrus
oleracea are palm species from the cerrados (savanna ecosystems from the
central-west and southeast Brazil). Examples of palm species from the caatinga
(semiarid ecosystem from northeast Brazil) are Copernicia prunifera (carnaba),
Syagrus coronata and Syagrus oleracea. In the Brazilian restingas (coastal
ecosystems with sandy and salty soils) occur Astrocaryum aculeatissum, Attalea
humilis, Bactris vulgaris and Syagrus romanzoffiana, among other palm species.
Many species of palm trees have enormous oleaginous potential. Not only the
palm oil tree, but also other palm species have long been important sources, such
as coconut (Cocos nucifera) and babau (Orbygnia spp.). The dry endosperm of
coconut may attain 60% of fat. The content of oil in the kernel of the babau
fruit may reach up to 70%. In addition, palm species may be prodigal fruit pro-
ducers, such as coconut, aa and babau. A single raceme of babau contains
normally around 250 fruits. These characteristics, allied to their capacity to grow
and reproduce well in tropical environments, demanding only little care, turn the
palm species interesting feedstock sources for biodiesel production in less-favored
regions of the Americas, Africa and Asia. They have been regarded as suitable
crops to be cultivated in degraded environments. As commented above, two palm
species (palm oil tree and tucum) have been recommended for cultivation in the
Brazilian Amazon, aiming biodiesel production.
In the biodiesel context, another point worth considering is the distribution of
fatty acids of lipids from palm species. They seem to contain low contents of poly-
unsaturated fatty acids. In both coconut fat and babau oil the main constituent
is lauric acid (dodecanoic acidsaturated C12), with very low or negligible con-
tents of linoleic and linolenic acids (C18 di- and triunsaturated acids, respectively).
Low content of polyunsaturated acids is a desirable characteristic in biodiesel
feedstock. The increase in the number of double bonds of fatty acids favors per-
oxidation processes during biodiesel burn, promoting the accumulation of carbon
deposits inside the engine cylinders (Anand et al. 2010). In addition, polyunsatu-
rated acids favor the emission of NOx (McCormick et al. 2001).
It is likely that lipids from many Arecaceae other than coconut and babau have
low content of polyunsaturated fatty acids. The distribution of fatty acids has been
shown to bear taxonomic meaning, i.e. closely related species tend to have simi-
lar distribution of fatty acids (Santos and Salatino 1998; Mayworm and Salatino
2002). Hence many palm species are expected to produce seed oils with low con-
tents of polyunsaturated fatty acids. Data on Table9.1 strengthen this hypothesis.
Among the 14 listed species, only 2 (Euterpe edulis and Oenocarpus bacaba)
have oils with high content of diunsaturated acid (linoleic). The oils from the other
species may be combined in two groups: (1) oils with predominance of palmitic
(saturated) and oleic (monounsaturated) acidsAcrocomia aculeata, Astrocaryum
vulgare, Bactris gasipaes, Elaeis guineensis, Euterpe oleracea, E. precatoria,
Mauritia flexuosa, M. vinifera, Oenocarpus bacaba and O. bataua; (2) oils with
predominance of lauric and myristic acids, both with saturated and medium length
carbon chainsAcrocomia aculeata, Astrocaryum murumuru, Cocos nucifera and
Orbygnia phalerata. Both species groups have low or negligible contents of the
polyunsaturated acids linoleic and linolenic (Table9.1). It is interesting to note in
Table9.1 that oils from palm species seem to have practically no fatty acids with
carbon chains longer than C18. Few species on the table have contents of oil lower
than 20%, and the oil yield of several species are considerably higher than the
yield of oleaginous species currently exploited (e.g. soybean: 20%). Taking also
into account the common high production of fruits, favorable perspectives of pro-
ductivity per hectare may be expected from the exploitation of palm species as
biodiesel sources.
Despite the many advantages and high potential of palm trees as biodiesel
feedstock, care should be taken during its exploitation. For example, it is neces-
sary to take into consideration the area to be deforested in order to avoid threats
to the preservation of plant and animal species. The palm oil boom production
in Indonesia has provided vital income to small-scale farmers, but certainly too
large extents of natural tropical forests have given over to palm oil production.
Table9.1Oil yield and distribution of fatty acids of lipids of Elaeis guineensis (palm oil tree) and Brazilian native species of palms (Arecaceae)
Species Oil yield Fatty acids (%)
(%) Caprilic Capric Lauric Myristic Palmitic Palmitoleic Stearic Oleic Vaccenic Linoleic Linolenic Araquidic
(8:0) (10:0) (12:0) (14:0) (16:0) (16:1) (18:0) (18:1) (18:1 (18:2) (18:3) (20:0)
cis11)
Acrocomia 53.0 5.0 50.9 13.1 7.6 3.0 17.9 2.5
aculeata
(Jack)
Lood. ex
Martius1
Astrocaryum 31.0 2.7 2.0 51.6 25.8 6.0 2.9 5.7 3.0 0.1 0.1
murumuru2
Astrocaryum 18.0 0.8 22.9 2.9 67.6 1.2
vulgare
Mart.3
Bactris 8.0 0.5 41.3 5.9 3.3 43.1 1.7 4.1 2.3
gasipaes
Kunth4
Cocos nucifera 27.7 5.8 4.8 49.1 21.8 8.4 2.8 6.1 1.2
9 Perspectives in Brazil of the Contribution of Palm Trees
L.5
Elaeis guineen- 62.0 0.3 31.1 8.8 49.6 9.2 0.8
sis Jacq6
Euterpe edulis 39.7 3.4 23.1 0.8 6.3 15.6 15.6 36.4 1.5
Martius7
Euterpe 31.0 0.07 0.1 26.2 4.9 1.8 52.0 3.4 7.8 0.6 0.1
oleracea
Martius8
Euterpe 29.6 0.3 14.8 0.4 0.3 80.6 3.1
precatoria
Martius9
147
(continued)
Table 9.1continued
148
Species Oil yield Fatty acids (%)

(%) Caprilic Capric Lauric Myristic Palmitic Palmitoleic Stearic Oleic Vaccenic Linoleic Linolenic Araquidic
(8:0) (10:0) (12:0) (14:0) (16:0) (16:1) (18:0) (18:1) (18:1 (18:2) (18:3) (20:0)
cis11)
Mauritia flexu- 11.2 0.1 17.3 2.0 73.3 2.4 2.2
osa L.10
Mauritia vinif- 66.0 0.03 0.1 16.8 0.3 1.8 76.1 4.9 1.0
era L.10
Oenocarpus 23.6 22.5 0.7 4.7 34.41 34.9 0.9
bacaba
Mart.11
Oenocarpus 28.6 15.8 1.2 2.3 73.5 5.4 0.7
bataua
(Mart.)
Burret.11
Orbignya 66.0 6.0 5.0 44.0 17.0 8.0 4.5 14.0 2.0
phalerata
Mart.12
1Beln-Camacho et al. (2005); 2Mambrim and Barrera-Arellano (1997); 3Schirmann et al. (2011); 4Clement et al. (1998); 5Bhatnagar et al. (2009); 6Monde
et al. (2009); 7Panza et al. (2009); 8Bora and Rocha (2004); 9Escriche et al. (2009); 10Silva et al. (2009); 11Unpublished data; 12Machado et al. (2006)
J. M. Meyer and A. Salatino
Tree burning and soil degradation, sometimes in carbon-rich peatlands, have given
off enormous amount of global-warming gases (Gilbert 2012). For this reason,
the choice of adequate localities for palm plantation in rainforest is important,
for example, avoiding marsh or wetland environments. Many internet sites have
blamed palm oil plantations in Indonesia for having caused the death of hundreds
of orangutans, a primate species already assumed as under serious risk of extinc-
tion. A possibility of cultivation in tropical forest with reduced or no risk of eco-
logical harm is the establishment of plantations in already degraded Amazonian
zones, such as has been done with black pepper (Piper nigrum) (Kato et al. 2001).
A seemingly interesting way of exploitation of native products is the harvest
of fruits in extractive regimen in preserved areas, created for this specific aim
(Mayworm et al. 2011). Extensive extractive areas have been established in sev-
eral locations of the Brazilian Amazon (Dubois 1996). The adequate conservation
of extractive areas depends on the sustainable exploitation of a high diversity of
species. Given the several favorable aspects of frequency of occurrence of palm
species in the Amazon, the commonly high productivity of fruits and the adequate
chemical profiles, the attainment of biodiesel feedstock in extractive areas is seem-
ingly promising from the economic, social and environmental viewpoints.
9.4Concluding Remarks
It is important to widen the exploitation of biodiesel feedstock toward a wider

spectrum of plant species. Little has been done in Brazil aiming the exploitation
of native species. Plants of Arecaceae are particularly interesting in this regard.
The introduction of products from native palm species may help minimizing the
food versus fuel dispute in biodiesel production (Lam et al. 2009): most biodiesel
feedstocks produced in Brazil are important edible products. The exploitation of
native palm species in Brazil may ameliorate the social-economic condition of
many people and help in the preservation of species and ecosystems, as far as pre-
cautions are taken and adequate and environmental friendly measures are put into
effect.
Acknowledgments The authors thank provision of funds by CNPq (Conselho Nacional do
Desenvolvimento Cientfico e Tecnolgico) and FAPESP (Fundao de Amparo Pesquisa do
Estado de So Paulo).
References
Agncia Nacional do Petrleo (ANP) (2012) Available at htpp://www.anp.gov.br. Accessed on

Oct 2012
Anand A, Ranjan A, Mehta PS (2010) Estimating the viscosity of vegetable oil and biodiesel
fuels. Energy Fuels 24:664672
Axelsson L, Franzn M (2010) Performance of jatropha biodiesel production and its environ-
mental and socio-economic impactsa case study. Chalmers, Department of Energy and
Environment, Gteborg, 79 (available at http://www.focali.se/filer/FRT%20201006%20-%20
Jatropha.pdf. Accessed on Jan 2013
Beln-Camacho DR, Lpez I, Garca D, Gonzlez M, Moreno-lvarez MJ, Medina C (2005)
Physicochemical evaluation of seed and seed oil of corozo (Acrocomia aculeata Jacq.).
Grasas Aceites 56:311316
Bhatnagar AS, Kumar PKP, Hemavathy J, Krishnae AGG (2009) Fatty acid composition, oxida-
tive stability, and radical scavenging activity of vegetable oil blends with coconut oil. J Am
Oil Chem Soc 86:991999
Bora PS, Rocha RVM (2004) Macaiba palm: fatty and amino acids composition of fruits. Ciencia
y Tecnologa Alimentaria 4:158162
Buckeridge MS, Salatino A (2010) Feedstock biochemistry applied to biofuels in Brazil. In:
Coser TR, Davis MJ (eds) Biofuels: reasonable steps towards a renewable energy source.
Fulbright Comission Brazil, Braslia, 1, pp 8589
Cintra R, Ximenes AC, Gondim FR, Kropf MS (2005) Forest spacial heterogeneity and palm
richness, abundance and community composition in Terra Firme forest, Central Amazon.
Braz J Bot 28: 7584
Clement CR, Aguiar JPL, Arkcoll DB (1998) Composio qumica do mesocarpo e do leo de
trs populaes de pupunha (Bactris gasipaes) do Rio Solimes, Amazonas, Brasil. Rev Bras
Fru 20:115118
Dias GLS (2007) Um desafio novo: o biodiesel. Estudos Avanados 21:179183
Dubois JCL (1996) Uses of wood and non-wood forest products by Amazonian dwellers.
Unasylva 186:815
Emerging Markets Online (2013) Biodiesel 2020: A global market survey, 2nd ed. Available at
http://www.emerging-markets.com. Accessed on Jan 2013
Environmental Protection Agency (EPA) (2002) Available at: http://epa.gov/air/caaac/mstrs/
korotney.pdf. Accessed on Oct 2012
Escriche I, Restrepo J, Serra JA, Herrera LF (1999) Composition and nutritive value of
Amazonian palm fruits. Food Nutr Bull 20:361365
European Biofuels (2013) Technology platforms. Biofuels in South America and Central
America. Available at http://www.biofuelstp.eu/s_america.html. Accessed on Jan 2013
Gilbert N (2012) Palm-oil boom raises conservation concerns. Nature 487 (available at
http://www.nature.com/news/palm-oil-boom-raises-conservation-concerns-1.10936.
Accessed on Jan 2013
Henderson A, Galeano G, Bernal R (1995) Field guide to the palms of the Americas. Princeton
University Press, Princeton
Kato AK, Albuquerque FC, Duarte MLR, Carvalho JEU, Hamada M (2001) Cultivo da Pimenta-
do-Reino sob Sistema Sombreado. Comunicado tcnico, Embrapa, http://www.infoteca.cnpti
a.embrapa.br/bitstream/doc/403378/1/ComTec55.pdf. Accessed on Jan 2013
Lam MK, Tan KT, Lee KT, Mohamed AR (2009) Malaysian palm oil: Surviving the food versus
fuel dispute for a sustainable future. Renew Sustain Energy Rev 13:14561464
Machado GC, Chaves JBP, Antoniassi R (2006) Composio em cidos graxos e caracterizao
fsica qumica de leos hidrogenados de coco babau. Ceres 53:463470
Mambrim MCT, Barrera-Arellano D (1997) Characterization of palm tree fruit oils from
Brazilian Amazonia region. Grasas Aceites 48:154158
Mayworm MS, Salatino A (2002) Distribution of seed fatty acids and the taxonomy of
Vochysiaceae. Biochem Syst Ecol 30:961972
Mayworm MS, Buckeridge MS, Marquez UML, Salatino A (2011) Nutritional reserves of
Vochysiaceae and potential economic uses. An Acad Bras Cienc 83:523531
McCormick RL, Grabowski MS, Alleman TL, Herring AM, Tyson KS (2001) Impact of biodiesel
source material and chemical structure on emissions of criteria pollutants from a heavy-duty
engine. Environ Sci Tech 35:17421747
Monde AA, Carbonneau M-A, Eymard-Duvernay S, Tiahou G, Badiou S, Adon B, Cristol J-P
(2009) Comparative study of fatty acid composition, vitamin E and carotenoid contents of
palm oils from four varieties of oil palm from Cte d Ivoire. J Sci Food Agric 89:25352540
Panza V, Pighin D, Linez V, Pollero RJ, Maldonado S (2009) Storage lipids and proteins of
Euterpe edulis seeds. Biocell 33:99106
Pike Research (2011) Biofuels markets and technologiesGlobal ethanol, biodiesel, and feedstock
markets: technology assessments, policy issues, key supply chain stakeholders and worldwide
supply and demand forecasts. p 144. http://www.pikeresearch.com. Accessed on Oct 2012
REN21 (2005) renewable energy policy network for the 21st Century. Renewables global status
report. http://www.ren21.net/Resources/Publications/REN21Publications.aspx. Accessed on
Oct 2012
Santos DYAC, Salatino A (1998) Fatty acids and seed oils of species of Diplusodon Pohl
(Lythraceae). Biochem Syst Ecol 26:109115
Schirmann GS, Rebollar PM, Miller PRM (2011) Aa na dieta brasileratradies regionais e
valor nutricional. VII Congresso Brasileiro de Sistemas Agroflorestais. 1922
Silva SM, Sampaio KA, Taham T, Rocco SA, Ceriani R, Meirelles AJA (2009) Characterization
of oil extracted from buriti fruit (Mauritia flexuosa) grown in the Brazilian Amazon Region. J
Am Oil Chem Soc 86:611616
Souza VC, Lorenzi H (2005) Botnica Sistemtica. Instituto Plantarum, So Paulo
Sze C, Whinihan J, Olson B, Schenk C, Sobotowski RA (2007) Impact of test cycle and biodiesel
concentration on emissions. SAE technical paper 2007-01-4040
Xue J, Grift TE, Hansen AC (2011) Effect of biodiesel on engine performances and emissions.
Renew Sust Energ Rev 15:10981116
Yazdani SS, Gonzalez R (2008) Engineering Escherichia coli for the efficient conversion of glyc-
erol to ethanol and co-products. Metabol Eng 10:340351
Zhuang D, Jiang D, Liu L, Huang Y (2011) Assessment of bioenergy potential on marginal land
in China. Renew Sust Energ Rev 15:10501056
Chapter 10
Xylan Biosynthesis in Plants, Simply
Complex
Ahmed Faik, Nan Jiang and Michael A. Held
AbstractXylans are major non-cellulosic polysaccharides in grasses and trees

and represent the third most abundant biopolymer on earth (after cellulose and
chitin). Xylans have important impacts on biofuel production because they are
contributors to plant biomass recalcitrance, yet plants deficient in xylans synthe-
sis grow abnormally. Therefore, deciphering the biochemical mechanisms of xylan
biosynthesis will undoubtedly contribute to identifying ways to improve biofuel
yields from plant biomass. Arabidopsis irregular xylem (irx) mutants have shown
that genes from GT43 and GT47 CAZy families encode proteins associated with
xylan biosynthesis. These genes are duplicated, have overlapping expression
patterns, and thus exhibit partial functional redundancy. However, genes from
one pair are incapable of complementing mutations in the other, suggesting that
their encoded proteins may function cooperatively in xylan synthase complexes
(XSCs), and recent work in wheat supports the existence of such XSCs. More
recent genetic studies in Arabidopsis suggest that xylan backbone elongation/syn-
thesis can be uncoupled from side chains additions to a certain extent. However,
what we still dont know is how xylan backbone synthesis is initiated and then
elongated by XSCs? And how many different XSCs does a plant employ to make
xylans? In this chapter, we will discuss what we know about xylan backbone ini-
tiation, elongation, and uncoupled substitution of the backbone. We will also dis-
cuss recent advances in the regulatory mechanisms of xylan synthesis.
Keywords Xylan synthesis Hemicellulose Arabidopsis Wheat Grasses

Cell wall Irregular xylem GT43 GT47 GT8 Glycosyltransferases
A. Faik(*) N. Jiang
Environmental and Plant Biology Department, Ohio University, Athens, OH 45701, USA
e-mail: faik@ohio.edu
M. A. Held
Department of Chemistry and Biochemistry, Ohio University, Athens, OH 45701, USA
154 A. Faik et al.
10.1Introduction
Plant lignocellulosic biomass consists mostly of plant cell walls and holds great
promise as a raw material for renewable biofuels. It is expected that manipula-
tion of the biosynthetic pathways of plant cell wall polysaccharides will lead to
improvements in plant biomass yield, digestibility, and energy recovery (conver-
sion rates). Xylans are the major non-cellulosic polysaccharides in plant ligno-
cellulosic biomass from grasses and trees and represent the third most abundant
polymer on earth (after cellulose and chitin). Xylan polymers are found mainly
in the primary cell walls of grasses and the secondary cell walls of grasses and
dicots. In lignocellulosic biomass, xylans interact directly with lignin and cel-
lulose and significantly contribute to its recalcitrance to enzymatic digestion
(saccharification). Overcoming this recalcitrance requires the inclusion of one
or more steps of biomass pretreatment (Faik 2013). Xylans are polymers made
mostly of pentoses (5-carbon sugars), and this adds another level of complex-
ity for their efficient fermentation, as most fermenting microorganisms cur-
rently used in the biofuel industry prefer hexoses (6-carbon sugars). For plants,
xylans in primary cell walls affect many aspects of plant growth and survival.
Any deficiency in xylan synthesis directly impacts plant growth and survival. It
seems that plants xylans have evolved many structural variations to fit numerous
functions in the wall, which would allow plant growth under various environ-
mental conditions. Thus, the designation one polymer for all seasons applies
to xylans. The structures of xylans vary depending upon species, tissues, or even
cells within the same tissue (Ebringerova et al. 2005). For instance, the structure
of xylan in cell walls of the starchy endosperm is different from that of the aleu-
rone or vascular walls. These structural variations suggest that plants have mech-
anistic and/or regulatory strategies to control xylan fine structure. Therefore,
deciphering the biochemical mechanisms of xylan synthesis will undoubtedly
contribute toward improving the quantity and quality of plant biomass for biofu-
els production. Xylan biosynthesis is far from understood and the process turns
out to be surprisingly difficult to grasp. The primary thrust of this chapter is to
critically analyze recent advances in the field of xylan synthesis and structure
leading to our current understanding of the biochemical mechanisms and regula-
tion of xylan synthesis in plant, and how these advances might improve biofuels
production. To start, we will review some of the most recent work related to this
topic in dicots and grasses and attempt to answer questions such as: How many
genes does a plant need to make xylans? And how can we reconcile the genetic
data with biochemical data that suggest the involvement of cooperative, multi-
protein complexes? Next, we will discuss the sophisticated, multi-level regu-
latory mechanisms that control xylan synthesis and secretion. Finally, we will
discuss how our current knowledge of xylan synthesis may be used to improve
plant biomass for biofuels, as well as aspects of xylan synthesis that are lagging
behind and deserve more attention.
10 Xylan Biosynthesis in Plants, Simply Complex 155
10.2Biochemical Mechanisms of Xylan Synthesis: So Many

Genes, So Little Known
Despite the increasing attention paid to the field of xylan biosynthesis, there is
limited progress on the biochemical side of the process. It is clear that plants have
developed a sophisticated biosynthetic mechanism with a complex regulation sys-
tem to control the fine structural details of xylan biosynthesis. This system likely
includes a large number of genes and adapted synthesis and secretion pathways.
In fact, we have learned (and are still learning) from genetics in Arabidopsis
that many genes are involved in glucuronoxylan (GX) biosynthesis in secondary
cell walls (in xylem and fiber cells). These genes are members of a large fam-
ily, called IRREGULAR XYLEM (IRX) genes, that also includes genes associ-
ated with cellulose synthesis (Turner and Somerville 1997; Brown et al. 2005).
Of particular interest are three glycosyltransferase (GT) families classified in the
CAZy database as GT8, GT43, and GT47. Interestingly, IRX genes from GT43
and GT47 families exist as duplicates, namely IRX10/IRX10-L, IRX14/IRX14-L,
FRA8(IRX7)/F8H(IRX7-L), IRX9/IRX9-L, and IRX15/IRX15-L, and exhibit par-
tial functional redundancy and overlapping expression patterns. This gene redun-
dancy is an indication of the importance of xylan synthesis to plant development
and survival. On the other hand, IRX genes from GT8 family dont show such
redundancy, which may suggest more specific roles limited to a particular tissue
types or developmental stages. It is important that our readers keep in mind that
the current genetic advances in xylan biosynthesis are limited to irx phenotypes
in Arabidopsis. These phenotypes occur in a specialized tissue and any analysis
of genetic data should be put in this particular context. Generalizations regard-
ing the mechanisms of xylan synthesis may not yet be possible with these data
alone. For example, the description of GX synthesis in secondary cell wall should
be understood under the physiological and cell biological context of that partic-
ular tissue. This resembles the fabled description of an elephant by blind men,
each one describing the elephant through limited touch perception. With this in
mind, the Arabidopsis irx mutants have revealed the involvement of many genes in
xylan biosynthesis, but questions about how these gene products interact to make
xylan polymers in primary and secondary cell walls remain. Nevertheless, the irx
mutants indicate that IRX genes impact GX synthesis during secondary cell wall
deposition to varying degrees. These impacts are described in Table10.1, which
will be discussed as we progress through the text.
Biochemical advances in grasses has revealed new aspects of
glucurono(arabino)xylan (GAX) biosynthesis, such as the presence of multi-
enzyme complexes and a cooperative biosynthetic mechanism that genetics alone
could not demonstrate. We recognize that these aspects may well be specific to
growing tissues in grasses. Nonetheless, this provides us the opportunity in this
chapter to compare and contrast data from Arabidopsis (on secondary cell wall)
with the data from grasses (on primary cell wall), and develop a bigger picture
of xylan synthesis in plants. Therefore, we will separate our discussion of the
156 A. Faik et al.
biosynthetic process of xylan biosynthesis into four sub-sections: II-1 xylan back-
bone initiation/termination, II-2 xylan backbone elongation, II-3 xylan backbone
additional decoration through an uncoupled mechanism, and II-4 xylan secretion/
delivery. These sections will be used to help integrate genetic and biochemical
advances into a more general mechanism that may be applied to both dicots and
grasses, as well as to different plant tissues.
10.2.1Xylan Backbone Inititiation/Termination

in Arabidopsis: Does Sequence 1 Have a Say?
It has been shown in dicots that GX polymers of the secondary cell wall have a tet-
rasaccharide (called sequence 1), 4--d-Xylp-(1,4)--d-Xylp-(1,3)--l-Rhap-(1,2)-
-d-GalpA-(1,4)-d-Xylp, located at the reducing end of their chains (Table10.1)
(Johansson and Samuelson 1977; Andersson and Samuelson 1983; Pena et al.
2007). It was hypothesized that sequence 1 is important for the initiation or ter-
mination of GX backbone elongation, based on the observation that this sequence
is absent in GX-deficient Arabidopsis mutants fra8(irx7), irx8, and parvus (Pena
et al. 2007; Lee et al. 2007). However the fact that fra8(irx7) and irx8 mutants can
still produce some GX (Lee et al. 2007), suggests that it is not absolutely required
for priming xylan backbone synthesis. Other irx mutants having shorter xylan back-
bone chains (such as irx9, irx10, and irx14), or double mutants having very little
GX (such as irx10/irx10L or irx14/irx14-L, Table10.1), still show the presence of
this oligosaccharide in their GX, which may suggest that somehow sequence 1 pre-
maturely terminates xylan backbone chain elongation (Brown et al. 2009; Wu et al.
2009, 2010). In support of this hypothesis, irx7 and irx8 mutants (both have GXs,
but lack sequence 1) have xylan backbone chains with increased length and het-
erodispersity (Brown et al. 2007; Persson et al. 2007; Lee et al. 2009) (Table10.1).
However, it is still unclear what mechanism allows a plant cell to sense the length
of xylan backbone and terminate the elongation. One possibility is that the addition
of sequence 1 occurs in one of the late Golgi compartments, while GX backbone
synthesis starts in early Golgi compartments. Thus, the distance traveled between
the two compartments could represent the time for xylan backbone elongation.
Defective synthetic machinery may have a slower synthesis rate to allow the pro-
duction of normal sized GX chains. It also appears that GlcA substitution does not
dictate the addition of sequence 1, as the Xyl to GlcA substitution ratio is unaltered
in all irx mutants (regardless the length of xylan backbone, Table10.1).
Taking all of these observations into consideration, we hypothesize that
sequence 1 may have an additional/alternative role in xylan synthesis in second-
ary cell walls. Specifically, sequence 1 might be involved in driving the secretion
of the newly synthesized GX to the cell surface. Secondary cell wall deposition
occurs at a specific developmental stage (e.g., after cell elongation is ceased), and
may require a rapid, signal-mediated secretion mechanism to deliver the massive
Table10.1Comparison of the effects of mutations in IRX7/IRX7-L, IRX9/IRX9-L, IRX10/IRX10-L, IRX14/IRX14-L, and IRX15/IRX15-L pairs along with
IRX8 and PARVUS genes on GX biosynthesis and phenotype of Arabidopsis plants
Mutants (acces- irx Xylan chain GX content com- Xyl substitution X6-dependent XylT Sequence 1 References
sion #) phenotype length pared to WT activity at reducing
end
fra8(irx7) Strong Increased and 59% Unaffected Unaffected Absent Pena et al. (2007),
(At2g28110) hetero-disperse Brown et al.
(2009), Lee et al.
(2009)
f8h(irx7-L) Mild Not known 20% Unaffected Unaffected Unknown Brown et al. (2009),
(At5g22940) Lee et al. (2009)
irx7/irx7-L Strong Reduced 85% Not known Unaffected Absent Wu et al. (2010),
Persson et al.
(2007)
irx10 (At1g27440) Mild Reduced 10% Unaffected Unaffected Present
10 Xylan Biosynthesis in Plants, Simply Complex
irx10-L No Unaffected 100110% Unaffected Unaffected Present Brown et al. (2009)

(At5g61840)
irx10 irx10-L Strong Not known Absent Decreased Present Wu et al. (2010)
irx9 (At2g37090) Strong Reduced 50% Unaffected Decreased Present Pena et al. (2007)
irx9-L Mild Not known 20% Unaffected Not known Present
(At1g27600)
irx9 irx9-L Strong Reduced 85% Unaffected Decreased Present
irx14 (At4g36890) Mild Reduced 50% Unaffected Decreased Present Wu et al. (2010)
irx14-L No Not known 20% Unaffected Not known Present Wu et al. (2010)
(At5g67230)
irx14 irx14-L Strong Not known Absent Decreased Present Wu et al. (2010)
irx8 (At5g54690) Strong Increased and 50% Unaffected Unaffected Absent Pena et al. (2007)
hetero-disperse
Parvus Strong Not known 50% Unaffected Unaffected Absent Lee et al. (2007)
(At1g19300)
157
(continued)
Table10.1continued
158
Mutants (acces- irx Xylan chain GX content com- Xyl substitution X6-dependent XylT Sequence 1 References
sion #) phenotype length pared to WT activity at reducing
end
irx15 (At3g50220) No Unaffected 25% Unaffected Unaffected Present Brown et al. (2011),
Jensen et al.
(2011)
irx15-L No Unaffected Unaffected Unaffected Unaffected Present Brown et al. (2011),
(At5g67210) Jensen et al.
(2011)
irx15/irx15-L Mild, Reduced 35% Unaffected Slightly decreased Present Brown et al. (2011),
uneven Jensen et al.
xylan (2011)
distribu-
tion)
A. Faik et al.
Fig. 10.1Model for sequence 1 as a secretion signal during GX synthesis in second-

ary cell wall. GX secretion would require a fast signal-mediated pathway from trans Golgi
network (TGN) to cell surface. Xylan backbone synthesis is initiated in cis Golgi network
(CGN) or cis-Golgi stacks and completed before reaching trans Golgi stacks or TGN, where
sequence 1 terminate xylan backbone elongation. Mutations in IRX9/IRX9-L, IRX10/IRX10-L,
and IRX14/IRX14-L genes would result in a defective xylan synthase complex with a slow elon-
gation process, and early termination by the transfer of sequence 1 onto GX with shorter chains.
Mutations in IRX7/IRX7-L, IRX8, and PARVUS genes would affect either the synthesis of
sequence 1 or its transfer onto GX, which would result in slower secretion to the secondary cell
wall and accumulation of GX with slightly longer backbone chains in the Golgi
amounts of GX polymer needed for building the secondary cell wall. The absence
of such a secretory mechanism would result in the accumulation of GX in the
Golgi of mature plant cells that usually lacking high vesicular trafficking to accom-
modate this need. In addition to this fast-delivery pathway for GX, a default secre-
tion pathway with a much slower rate may exist for other secretory cargo including
some GX (Fig.10.1). In this model, sequence 1 would be synthesized in the trans-
Golgi or trans Golgi network (TGN) by a sequence 1 synthase complex (indi-
cated as in Fig.10.1) that can also act as a carrier. This sequence 1 synthase
complex would transfer sequence 1 to the newly synthesized GX, which would
160 A. Faik et al.
facilitate its secretion, possibly with the help of a protein or lipid receptor that
recognizes sequence 1 as a secretion signal. In this model, xylan backbone syn-
thesis would be initiated early in the Golgi (e.g., cis Golgi) by xylan synthase com-
plexes (indicated as red in Fig.10.1) and completely decorated before reaching
the late Golgi compartments (i.e., trans-Golgi stacks or TGN). The whole GX-
sequence 1 polymer can then be recognized by its putative receptor, packaged
in vesicles (most likely of clatherin type), and then delivered to the cell surface
(Fig.10.1). This model is in agreement with the current genetic data. For example
mutations that yield defective GX synthases (irx9, irx10, and irx14, indicated as
red in Fig.10.1) would produce GX with shorter xylan chain lengths (at much
slower rate) compared to wild type. These shorter GX chains would still be termi-
nated by sequence 1 in the trans-Golgi or TGN and delivered to the cell surface
(Fig. 10.1). In the case of irx7, irx8 and parvus mutants, defects in the transfer
reaction of sequence 1 onto the newly synthesized GX or in sequence 1 synthase
complex itself (indicated as in Fig.10.1) would result in slower secretion of
GX to the cell wall and possibly the accumulation of large amounts of GX in the
Golgi. Another possibility is that the absence of sequence 1 may delay the release
of newly synthesized GX from xylan synthase complexes. This would result in
the accumulation of GX having longer xylan backbone chains in trans-Golgi or
TGN and could explain the longer xylan backbone chains observed in GX from
irx7 and irx8 mutants (Fig.10.1). We can imagine that the accumulation of GX
in trans-Golgi or TGN might trigger a reduction in GX biosynthesis by a nega-
tive feedback mechanism (yet to be discovered). Signal-mediated secretion would
appear to be specific to GX in secondary cell wall of vascular tissues, as cellulose-
deficient Arabidopsis mutants (such as irx3) do not show a decrease in GX con-
tent (Brown et al. 2011). The same situation was observed in irx4, an Arabidopsis
mutant defective in lignin biosynthesis (mutation in a cinnamoyl CoA reductase
gene), which exhibits collapsed xylem and altered plant development, but retains
normal levels of GX (Jones et al. 2001).
Another possible function of sequence 1 could be for the integration of GX into
the secondary cell wall. For example, sequence 1 may be needed to attach GX at
specific sites in the cell wall (i.e., to lignin or pectins), which would also drive GX
secretion. Several types of covalent bonds that connect xylans to lignin (and other
cell wall polymers) within the secondary cell wall have been documented (Joseleau
et al. 1992). These covalent bonds include glycosidic linkages between free C-1
positions (reducing end) of Xyl residues and p-Coumaric acid of lignin; benzyl
ester linkages between the side chains of GlcA and/or Ara residues and lignin (via
ferulic or coumaric acids); and ether linkages between lignin and either Ara or Xyl
residues of side chains. Most of these linkages are susceptible to alkaline treatment,
but can be identified by enzymatic treatment of xylans (Joseleau et al. 1992).
So far, there is no experimental evidence of the presence of sequence 1 in mono-
cots (Fincher 2009). However, work by Zeng et al. (2010) showed that a puri-
fied wheat xylan synthase complex (XSC) could produce in vitro a GAX-like
polymer that releases two main oligosaccharides (called peak I and peak II) after
digestion with endoxylanase III. Structural analysis of peak I suggested that it was
an oligosaccharide with a degree of polymerization (DP) of 8 made exclusively of

Xyl and Ara residues in a ratio of 3:1, respectively. Peak II, on the other hand was not
fully characterized, but contained mainly Xyl and GlcA (with very little Ara). This
uncharacterized oligosaccharide could be a variation of sequence 1 needed for initia-
tion/termination/secretion of GAX biosynthesis in monocot primary cell walls. The
fact that grass homologs of Arabidopsis IRX7, IRX8, and PARVUS genes (required for
synthesis of sequence 1) are not highly expressed in tissues that produce large amount
of AX in grasses strongly suggests that xylans in primary cell walls dont need
sequence 1 for chain elongation termination and secretion. Similarly, low expression
of Arabidopsis IRX7, IRX8, and PARVUS genes was observed in transcriptional pro-
filing studies carried out in psyllium (Plantago ovata Forsk) seed mucilage, a tissue
that is rich in xylans and primary cell walls (Jensen et al. 2011). This is in agreement
with the fact that psyllium mucilage xylans lack the terminal sequence 1 oligosac-
charide (Jensen et al. 2011). Our hypothesis is that a GlcA-rich oligosaccharide,
with a structure yet to be determined, may act as a terminator in xylan biosynthesis
in grasses. In fact, intact wheat microsomes generated only short GAX polymers in
vitro, having a DP ranging from 50 to 80 (Zeng et al. 2008). It would be interesting
to test the effect of sequence 1 on GAX synthesis in secondary cell walls of grasses.
Current data suggest that xylan chain termination mechanisms might be different
between grasses and dicots. For example, grass homologs of Arabidopsis IRX7, IRX8,
and PARVUS genes, which are required for synthesis of sequence 1, are not highly
expressed in tissues that produce large amount of AX in grasses.
It is becoming evident that dissecting the steps of GX biosynthesis within
Arabidopsis Golgi apparatus will be critical toward a full understanding of the
biosynthetic process. It will be important to determine where sequence 1 is syn-
thesized and transferred onto GX, and what the exact roles of IRX7, IRX8, and
PARVUS are in the process. Developing a biochemical assay for GX synthesis
in secondary cell walls in Arabidopsis will be a major breakthrough for testing
the biochemical function of these GTs. More importantly, it will be important to
assess the role of sequence 1 in the secretion of GX to the cell surface, and iden-
tify possible GTs that catalyze the transfer of sequence 1 onto GX. Although these
types of experiments are difficult to implement, any information gained from them
would help tremendously advance the field of xylan biosynthesis. Plant cell wall
polysaccharide secretion is still lagging behind and more efforts are needed to
understand this area. Our proposed model for sequence 1 in GX secretion provides
a framework to design experiments to answer specific questions in this field.
10.2.2Xylan Backbone Elongation: Cooperative Mechanism

and Core Xylan Synthase Complexes
Several observations provide direct evidence that xylan backbone synthesis and
elongation is under the control of more than one protein. For example, genes
162 A. Faik et al.
from one IRX pair are incapable of complementing mutations in the other pairs
(i.e., ixr9/irx9-L double mutant is not complemented by IRX10, IRX14, F8H, or
FRA8), and mutations in IRX genes have additive effects on irx phenotype and
plant growth (Table10.1). Other observations provide indirect evidence that the
IRX proteins function cooperatively in XSCs. For instance, the degree of sub-
stitution of the xylan backbone with GlcA residues is not affected by mutations
in IRX genes regardless of the length of the xylan backbone chain (Table10.1).
This is only possible if xylan backbone elongation is coupled with the addi-
tion of GlcA side chains via the cooperative action of multi-protein complexes.
Although our work in wheat supports the existence of XSCs and provides evi-
dence of the involvement of a cooperative mechanism for elongating of GAX-like
xylan backbone polymers (Zeng et al. 2008, 2010), there is no equivalent GAX
synthesis assay in mature Arabidopsis plants, making it difficult to translate data
from etiolated wheat seedlings (grass) to mature Arabidopsis plants (dicots). To
gain insights into GAX biosynthesis in dicots, we sought to test if etiolated wild
type Arabidopsis seedlings have GAX synthesis activity similar to that observed
in etiolated wheat seedlings. To do so, Golgi-enriched microsomes were prepared
from 5 to 7-day old etiolated Arabidopsis seedlings and tested for GAX synthesis
activity as described in Zeng et al. (2010). This previous work showed a stimula-
tory effect of UDP-Xyl on the incorporation of [14C]GlcA, from UDP-[14C]GlcA,
into ethanol-insoluble products using the wheat microsomes. Similarly, intact
Arabidopsis microsomes showed substantial [14C]GlcA incorporation in pres-
ence of UDP-Xyl, and no [14C]GlcA incorporation in the absence of UDP-Xyl
(Fig.10.2). Monosaccharide composition analysis of the [14C]-product generated
by Arabidopsis microsomal membranes in presence of UDP-[14C]GlcA, UDP-
[14C]Arap, and UDP-[14C]Xyl was then performed by acid hydrolysis (2M TFA,
1h at 120C) of the [14C]-product and fractionation by High pH anion exchange
chromatography (HPAEC) on a CarboPac PA20 column (Dionex). Our data show
that ~70% of the [14C]-radiolabel co-elutes with Ara, Xyl, and GlcA in a ratio
Xyl:Ara:GlcA of 9:0.5:1 (according to UDP-sugars specific [14C]-radioactivity),
suggesting less Ara incorporation (Fig.10.2).
General conclusions from these preliminary data are (1) Arabidopsis micro-
somes, like in wheat, also require the presence of both UDP-Xyl and UDP-
GlcA to produce a GAX-like polymer. Although the incorporation of [14C]GlcA
was ~10 times lower compared with wheat microsomes, GAX synthesis activity
was clearly present. This was expected, as GAX is only a minor constituent in
Arabidopsis primary cell walls compared to wheat; (2) GAX-like polymers pro-
duced by Arabidopsis microsomes have a ratio of Xyl:GlcA of 9:1, which is simi-
lar to the ratio of 8:1 observed in GX from the secondary cell walls of Arabidopsis
(Pena et al. 2007); and (3) like wheat, Arabidopsis uses a cooperative mechanism
to simultaneously incorporate GlcA and Xyl, which also suggests conserved
mechanisms of xylan synthesis in primary cell walls. It will be important to inves-
tigate GAX content and structure in cell walls from etiolated Arabidopsis seed-
lings to confirm the presence of Ara residues in xylan. Although GX polymers
from secondary cell walls have been shown to be devoid of Ara, earlier studies
Fig.10.2Analysis of glucurono(arabino)xylan (GAX) synthase activity in etiolated wild type

Arabidopsis seedlings grown on vertical plates. The activity was monitored via transfer of
[14C]GlcA from UDP-[14C]GlcA into ethanol-insoluble products in presence of Golgi enriched
micrsomal preparations (~0.2mg proteins from 5 to 7-day old seedlings). [14C]GlcA incorpora-
tion was measured in the presence (black bar) or absence (white bar) of UDP-Xyl as described in
Zeng et al. (2010). For product analysis all sugars were [14C]-radiolabeled and the [14C]-products
formed were acid hydrolyzed (2M TFA, 120C, 1h), and the monosaccharides released were
fractionated by High pH anion Exchange Chromatography (HPAEC, Dionex). The elution of
[14C]-radiolabeled material was monitored by counts (cpm) using a scintillation counter. These
analyses were conducted in triplicate
on primary cell walls in Arabidopsis suggested the presence of small amounts of

GAX (~4%) (Zablackis et al. 1995), which is comparable to most dicot plants
(Darville et al. 1980; Scheller and Ulvskov 2010).
Genetic analyses in Arabidopsis suggest that xylan backbone elongation of
GX synthesis is tightly linked to the addition of GlcA side chains. Unfortunately,
there is no experimental data from Arabidopsis irx mutants to directly support
the involvement of a cooperative mechanism and/or a multi-protein complex
(such as XSC from wheat). Purified wheat XSCs seem to contain only orthologs
to Arabidopsis IRX-L proteins (no IRX orthologs were present in the com-
plex), namely, TaGT43-4 (putative ortholog of IRX14-L) and TaGT47-13 (puta-
tive ortholog of IRX10-L) (Zeng et al. 2010, Faik unpublished data). This finding
raises several important questions related to xylan backbone elongation: (1) do
all IRX and IRX-L proteins interact with each other to form core XSCs? and
(2) how many of different XSCs does a plant need to make xylans? Our hypoth-
esis is that, like in the wheat core XSC, Arabidopsis IRX7/IRX7-L, IRX9/IRX9-L,
IRX10/IRX10-L, and IRX14/IRX14-L pairs form several core XSCs by pairing
partners from GT43 and GT47 families. For example, four putative core XSCs are
possible just from the combinations of IRX10/IRX10-L and IRX14/IRX14-L pairs
([IRX10-IRX14]; [IRX10-IRX14-L]; [IRX10-L-IRX14]; and [IRX10-L-IRX14-L]
complexes), and another four putative core XSCs could be formed from the
IRX9/IRX9-L and IRX7/IRX7-L pairs, generating a total of eight possible core
XSCs. Experimental data will be needed to substantiate the number of core XSCs
in Arabidopsis.
164 A. Faik et al.
The question then becomes, why have so many core XSCs? And how are they
functionally different? One possible answer is that some of these core XSCs are
specialized for certain tissues or at certain developmental time points (e.g., for
making xylans in primary vs. secondary cell wall). Since a wheat core XSC was
purified from etiolated seedlings rich in primary cell wall, it is reasonable to sug-
gest that some of these XSCs are involved in xylan synthesis in primary cell walls
(i.e., GAX and AX), while others have major roles in producing xylans for sec-
ondary cell walls (i.e., GX), reminiscent of the unique rosette terminal complexes
employed for primary and secondary cell wall biosynthesis. Nonetheless, the rea-
son for a plant to have specialized core XSCs for each type of cell wall is puzzling.
10.2.3Substitution of the Xylan Backbone: Uncoupled Versus

Coupled Addition of Side Chains
Recently, work by Anders et al. (2012) showed that silencing two -(1,3)-AraT
genes (XAT1 and XAT2, members of the GT61 family) did not affect the number
of substituted Xyl residues in AXs from the endosperm of wheat. It is known that
Ara residues in endosperm AXs are either -(1,2)- or -(1,3)-linked to the xylan
backbone, and that some Xyl residues can be di-arabinosylated with both link-
ages. Thus, there are technically four types of -AraTs that may be required for
arabinosylation of AXs in the wheat endosperm: -(1,3)-AraTs and -(1,2)-AraTs
that would add Ara onto unsubstituted Xyl residues of the backbone, and -(1,2)-
AraTs or -(1,3)-AraTs that would transfer Ara onto mono-substituted -(1,3)-
Ara or -(1,2)-Ara-linked Xyl residues, respectively to form di-substituted Xyl
residues.
The findings by Anders et al. (2012) have important implications regarding AX
synthesis in grasses. First, it suggests that despite the absence of -(1,3)-linked
Ara, xylan backbone elongation is still linked to the incorporation of Ara (mirror-
ing the situation of GlcA and Xyl incorporation in GX). Second, it also suggests
that -(1,2)-AraT activity, but not -(1,3)-AraT activity, may work cooperatively
with xylan synthase to synthesize AX polymer. Interestingly, the proteomics stud-
ies described by Zeng et al. (2010) did not identify any members of the GT61 fam-
ily associated with fractions enriched in wheat GAX synthase activity, and this
observation was confirmed by further proteomics analysis of the affinity-purified
GAX synthase complex (Faik and Jiang unpublished data). Thus, we are tempted
to conclude that the AraT detected with the purified wheat GAX synthesis activ-
ity is an -(1,2)-AraT activity. This -(1,2)-AraT may belong to a GT family
other than GT61. Members of the GT47 family could be candidates for this type
of -(1,2)-AraT activity. It will be necessary to provide experimental data to sup-
port this hypothesis either by directly testing the enzyme activity of the candidates
or through over-expressing them in Arabidopsis and analyzing xylan structures
from transgenic plants. Together, these observations still support a cooperative
mechanism for AX biosynthesis in wheat, but subsequent substitutions with Ara,

acetyl groups, or GlcA could be uncoupled from xylan backbone elongation.
The situation is more complicated for GX synthesis in secondary cell walls.
A recent work by Mortimer et al. (2010) showed that two Arabidopsis genes,
GUX1 and GUX2 (both member of the GT8 family), were responsible for the
addition of GlcA side chains onto GX in secondary cell wall. Interestingly, gux1
gux2 double mutant plants have xylan contents and xylan chain lengths similar to
wild type, yet have almost no detectable GlcA residues. Additionally, these mutant
plants have no detectable growth phenotype. Although this finding is difficult to
reconcile with a cooperative mechanism of the XSC complex, it has some impor-
tant implications toward understanding the GX biosynthetic mechanism. First,
the data show that somehow xylan backbone elongation can be uncoupled from
GlcA substitution during GX synthesis. A possible explanation is that GX from
gux1 gux2 double mutant plants may still have low amounts of GlcA not detect-
able by the carbohydrate gel electrophoresis (PACE) technique. Although the
PACE method is sensitive (detection of <1picomol of released sugars) (Goubet
et al. 2002), it also relies on derivatization of the reducing ends of sugars with
fluorophores, which could be a limiting step if the efficiency is low. In fact,
MALDI-TOF MS analysis of oligosaccharides from GX prepared from gux1
gux2 double mutant plants still shows a small peak at m/z 963.50 corresponding
to MeGlcA(Xyl)4 (Mortimer et al. 2010). A second possible explanation is that
other enzymes may have taken the place of the missing GlcAT in the XSC allow-
ing xylan backbone elongation to occur with low amounts of GlcA side chains
incorporation. An acetyltransferase (AcetylT) that adds acetyl groups to the C-6
positions of Xyl residues could be that enzyme. Lee et al. (2011) identified four
Arabidopsis REDUCED WALL ACETYLATION (RWA) genes (called RWA1,
RWA2, RWA3, and RWA4) required for xylan acetylation in the secondary cell
wall. When the four RWA genes were mutated, a significant reduction in second-
ary cell wall thickening was observed. Although mutant plants showed a milder
irx phenotype, there was no reduction in the amount of xylan Lee et al. (2011).
The authors did not analyze the xylan chain length from rwa1/2/3/4 mutant plants.
Although the mechanisms of xylan O-acetylation (or any plant cell wall polymer)
are not known (Gille and Pauly 2012), studies in mammalian systems indicate that
O-acetylation of sialic acid takes place in the lumen of the Golgi using acetyl-CoA
as a donor substrate to form acetylated intermediates (Higa et al. 1989). The exist-
ence of a GX polymer with low GlcA content and higher acetylation has been
documented in many species. For example, Goncalves et al. (2008) isolated a GX
polymer from hybrid Paulownia elongata/Paulownia fortunei (a deciduous tree)
that has much less GlcA substitution (Xyl:GlcA ratio of 20:1), but much higher
acetylation (almost 50% of Xyl are substituted with acetyl group), which may
suggest that AcetylT can substitute for GlcAT in a XSC. From these observa-
tions, we propose a model for xylan backbone elongation whereby all xylan types
are produced by core XSCs that elongate xylans by a coupled mechanism. These
XSCs are modular, depending of the species and tissues, and can be substituted
with alternate enzymes to overcome defects in xylan biosynthesis. The central
166 A. Faik et al.
Fig.10.3Model for synthesis of various types of xylans (GAX, GX, and AX) in primary and
secondary plant cell walls. Xylan synthase complexes (XSCs) are modular with a core complex
formed by a pair of proteins from IRX and IRX-L sets (see text). These cores XSCs are
responsible for the synthesis of the xylan backbone coupled with simultaneous incorporation of
some GlcA and/or Ara side chains. Depending of the degree of substitution, additiona enzymes
such as GUX1, 2 (GT8), XAT1, 2 and XAX1 (GT61), or RWA1,2,3, and 4, can be involved in
adding GlcA, Ara, Xyl, or acetyl groups, respectively, to the backbone in an uncoupled man-
ner. The mechanism of ferulic acid incorporation onto Ara or Xyl-Ara side chains by a family of
putative feruloyltransferases (Pfam PF02458) is still unknown
core of these XSCs is composed of members of the IRX and IRX-L proteins.
In this model (Fig.10.3), additional enzymes/GTs can work independently from
XSC (in an uncoupled manner), which would provide flexibility and adaptability
of the biosynthetic machinery for the production of many xylan types.
10.2.4Secretion/Deposition of Xylans to the Cell Wall
Plant cell wall polysaccharide secretion/deposition to the cell wall is still lagging
behind other aspects of cell wall biology, and certainly more efforts are needed to
advance this area of research. Our proposed model for sequence 1 in GX secretion
during secondary cell wall deposition would give a framework to design exper-
iments specifically to answer questions such as: Which proteins are involved in
recognizing sequence 1 in endomembrane system? Or which GTs catalyze the
transfer of this sequence 1 onto GX before secretion/deposition?
The recent characterization of two new irx mutants in Arabidopsis DUF579
genes (irx15 and irx15-L, Table10.1) start to shed the light on the mechanisms of
secondary cell wall deposition (Brown et al. 2011; Jensen et al. 2011). The double
mutant irx15 irx15-L plants have a mild irx phenotype with a decrease of up to
50% in GX content in stem secondary cell walls, with cellulose contents simi-
lar to many other GX-deficient mutants, see Table10.1. However, what is most
interesting about this double mutant is the uneven deposition of their secondary
cell walls (compared to the smooth and even deposition of secondary cell walls in
other irx mutants). Thus, it was proposed that IRX15 and IRX15-L proteins might
have a role in the transport/delivery of polysaccharides to the cell wall (Brown
et al. 2011). Although this proposed function needs experimental confirmation,
there is indirect evidence that supports it. For example, one might expect the
IRX15 and IRX15-L genes to be highly expressed in tissues that are actively syn-
thesizing and secreting xylans and this is indeed true in the case of xylem/vessel
tissues, and the mucilaginous layer (called husk) of Psylium seeds, which contains
more than 60% dry weight in heteroxylans (Fischer et al. 2004) and is a devel-
oping tissue actively synthesizing and secreting xylans. Also, structural analysis
of a DUF579-containing protein (YPL225W) from S. cerevisiae showed that it
contains a coiled-coil domain necessary for proteinprotein interactions, and
can interact with components of the cytoskeleton (i.e., ARP23 complex) and the
endomembranes (Costanzo et al. 2010; Brown et al. 2011). Therefore, it is pos-
sible that the DUF579-containing IRX15 and IRX15-L proteins may have the
capacity to interact with cytoskeleton and participate in GX/polysaccharide traf-
ficking. It is also possible that IRX15 and IRX15-L play a role during the bio-
synthesis of GX. For example, they could act as polysaccharides chaperones to
maintain GX in a soluble form during xylan backbone elongation and delivery to
the cell surface. Their putative role as polysaccharide chaperones could explain the
shorter xylan chain lengths observed in the irx15 irx15-L double mutant plants,
since the absence of these proteins would result in GX insolubility after a certain
length. This would cause a reduction in the rate of synthesis by xylan synthase
complexes, and aggregations of newly synthesized GX in the Golgi. The fact that
X6-dependent XylT activity and the ratio Xyl to MeGlcA are unaffected in the
irx15 irx15-L double mutant suggests that the xylan synthase complexes are func-
tional in this mutant (Brown et al. 2011) (Table10.1). Newly synthesized GX in
the irx15 irx15-L double mutant would still be decorated with sequence 1 and
delivered to cell wall, however, the absence of IRX15 and IRX15-L proteins may
restrict vesicular trafficking to a certain area of the cell surface through smaller
vesicles that are delivered in a patchy manner. Another possibility is that IRX15
and IRX15-L may contribute indirectly to GX solubility in the Golgi by playing
a role in the O-acetylation of xylan backbone. It is known that most hemicellu-
loses, including GX, are O-acetylated (Kiefer et al. 1989; Carpita 1996; Scheller
and Ulvskov 2010; Gille et al. 2011). In this scenario, IRX15 and IRX15-L could
stabilize GX-specific acetyltransferases (i.e., reduced wall acetylation proteins:
RWA1, RWA2, RWA3, and RWA4) on the newly synthesized GX. It would be
interesting to know whether RWA proteins interact with IRX15 and/or IRX15-L.
The Arabidopsis At1g33800 gene has been recently characterized and shown to
encode a 4-O-methyltransferase specific to GX (GXMT) (Urbanowicz et al. 2012).
This protein has also a DUF579 domain but clusters phylogenetically in a different
group from IRX15 and IRX15-L sharing less than 30% identity at the amino acid
168 A. Faik et al.
level. Thus, determining the biochemical function of IRX15 and IRX15-L will be
necessary before a clear conclusion regarding the roles of these proteins can be
drawn.
10.3Xylan Synthase Complexes: Primary Versus

Secondary Cell Wall
According to studies from Arabidopsis irx mutants (summarized in Table10.1),

not all IRX genes have equal importance in GX synthesis in the secondary cell wall
of mature Arabidopsis plants. For instance, while GX synthesis is not abolished in
the irx9/irx9-L double mutant, both the irx14/irx14-L and the irx10/irx10-L double
mutant make very little GX, indicating that IRX10/IRX10-L, and IRX14/IRX14-L
pairs have more important roles in secondary cell wall GX biosynthesis than the
IRX9/IRX9-L pair elongation (Wu et al. 2009, 2010). Furthermore, analysis of pro-
moter-GUS constructs of IRX genes expressed in Arabidopsis indicate that IRX7,
IRX9, IRX14, and IRX14-L are predominantly expressed in the central stele of
root tissues which are rich in secondary cell wall, while IRX9-L and IRX7-L are
expressed in the peripheral cell layers (tissues that are not making any secondary
cell walls) as well as in central stele of root (Wu et al. 2009). Similarly, vascular
tissues of Arabidopsis leaves showed restricted expression of IRX7, IRX9, IRX14,
and IRX14-L, while IRX9-L and IRX7-L are expressed throughout leaf tissues
(Wu et al. 2009). Additional promoter analyses using yellow fluorescence protein
(YFP) fusions and confocal microscopy support the general conclusion that IRX
genes are associated with the synthesis of secondary cell walls of xylem, while
IRX-L genes are weakly expressed in xylem yet their expression seems to be
more widespread (Wu et al. 2009). These observations give support to the hypoth-
esis that xylan biosynthesis in primary and secondary cell walls in Arabidopsis
may involve different sets of genes or XSCs, as was shown for cellulose synthesis
(for review see Joshi and Mansfield 2007; Somerville 2006). Confirmation of this
hypothesis may require purification of functional GAX synthase complexes from
etiolated Arabidopsis seedlings (rich in primary cell walls). In fact, all the con-
clusions regarding Arabidopsis IRX and IRX-L genes involved in GX biosyn-
thesis (Table10.1) are based on experimental work in the stems of mature plants
making secondary walls. Unfortunately, little work has been done on xylan synthe-
sis and structure in etiolated Arabidopsis seedlings.
Etiolated seedlings are comprised mostly of rapidly expanding mesocotyl (in
grasses) and hypocotyl cells (in dicots). These cells are rich in primary cell wall
material and therefore offer an excellent experimental model to study the biochem-
istry, enzymology, and regulation of GAX synthesis and for testing the genes asso-
ciated with the process. One would expect to see a phenotype in seedlings having
mutations in genes important for GAX synthesis in primary cell walls. In an attempt
to address this hypothesis and determine the role of IRX and IRX-L genes in
Fig.10.4Analysis of the growth of Arabidopsis irx7, f8h(irx7-L), irx9, irx9-L, irx10, irx10-L,
irx14, irx14-L, irx8, and parvus mutants on vertical plates in dark (panels a and b) or under light
(panels c and d). Seeds were surface sterilized and cold treated for two days before planting.
Pictures were taken 57days after germination
xylan synthesis in primary cell walls, we monitored the growth of Arabidopsis irx
mutants (irx7, irx7-L, irx9, irx9-L, irx10, irx10-L, irx14, irx14-L, irx8, and parvus)
on vertical plates under light and dark conditions. When Arabidopsis irx mutant
seedlings were grown in the dark (etiolated), hypocotyl elongation seemed to be
affected by mutations in IRX9-L, IRX10-L, or IRX14-L genes, as the seedlings from
these mutants were 3050% shorter than wild type (Fig.10.4a). On the other hand,
mutations in IRX7, IRX9, IRX10, or IRX14 had only minor effects on hypocotyl
170 A. Faik et al.
elongation (~10% shorter compared to wild type, Fig.10.4a). Perhaps the most dra-
matic effect observed was the very delayed germination of seeds from f8h(irx7-L)
mutants (Fig.10.4a). Surprisingly, when the same irx mutants were grown under
lighted condition, no significant differences in hypocotyl lengths were observed
among all seedlings of irx mutants and wild type (Fig.10.4b). However, a reduc-
tion in root length by ~50% was observed for irx14 and f8h(irx7-L) compared to
wild type (Fig.10.4b). Seeds from f8h(irx7-L) mutant seem to have shorter ger-
mination delay under light conditions. It is worth noting that both irx8 and parvus
mutants did not show any significant growth differences compared to wild type
under light and dark conditions (Fig.10.4c and d). This result is expected since both
IRX8 and PARVUS genes are required for sequence 1 synthesis and the secretion of
GX during secondary cell wall deposition. Sequence 1 is absent in other xylans such
as GAX and AX. This preliminary finding supports the notion that sequence 1 may
not be required for GAX synthesis in growing tissues with primary cell walls.
Taken together, these observations indicate that mutations in both sets of xylan
synthesizing genes, IRX and IRX-L from GT43 and GT47 families, affect
hypocotyl elongation of Arabidopsis seedlings, but IRX-L genes have a stronger
effect, which is the reverse situation when Arabidopsis is grown on soil and mak-
ing stems with secondary cell walls (Wu et al. 2009, 2010). Although this finding
gives additional support to the conclusion that IRX and IRX-L genes have
varying importance in xylan biosynthesis in primary and secondary cell wall, fur-
ther analyses will be needed to determine the exact roles of each of these genes on
plant development. We propose that IRX genes are more important in GX bio-
synthesis in secondary cell walls, while the IRX-L genes are more critical for
xylans synthesis in tissues rich in primary cell walls. The general distribution trend
of IRX and IRX-L gene expression within Arabidopsis body would look like
two opposite gradients as depicted in Fig.10.5. Figure10.6 summarizes expression
profiles of IRX7/IRX7-L, IRX9/IRX9-L, IRX10/IRX10-L, and IRX14/IRX14-L pairs
in Arabidopsis roots using publicly available microarray data (http://bar.utoronto.
ca/efp/cgi-bin/efpWeb.cgi; Brady et al. 2007; Winter et al. 2007). These expres-
sion data are in good agreement with our hypothesis and show that, in general, the
expression of IRX genes are more associated with tissues making secondary cell
walls, while IRX-L genes are more associated with growing/dividing tissues mak-
ing primary cell walls. This differential role (and expression) of two very close genes
highlights another level of regulation of xylan synthesis in dicots.
10.4Regulation of Xylan Biosynthesis
In theory, cell wall polysaccharide biosynthesis can be regulated at several levels,

including gene expression, the abundance and stability of mRNAs, the production and
stability of the protein (translation and post-translational modifications), the effects of
protein complex formation, and the secretion/delivery of newly synthesized polysac-
charides to the cell surface and their integration into the cell wall. We are just starting
Fig.10.5Schematic
presentation of the general
distribution of IRX
and IRX-Like genes in
Arabidopsis stems. Primary
and secondary growth is
indicated by arrows. The
expression of IRX genes
increases with secondary
growth, while the expression
of IRX-Like genes
increases in tissues with
primary growth
to elucidate the roles of some transcription factors (TFs) and small RNA regulators
of cell wall biosynthesis. We are also starting to identify some of the proteinprotein
interactions needed in assembling functional polysaccharide synthase complexes.
However, we are still far from developing a larger picture of the regulatory landscape
associated with the synthesis of any given cell wall polymer. Xylan biosynthesis is no
exception, considering their importance in plant growth and survival.
10.4.1Regulators of IRX Gene Expression: Xylan Versus

Other Cell Wall Polymers
Currently there are about 20 known proteins that constitute the Arabidopsis TF reg-
ulatory network for cell wall biosynthesis. Elucidating the transcriptional regulatory
mechanisms of cell wall synthesis is hampered by the fact that promoter sequences
172 A. Faik et al.
Fig.10.6Expression profiles of IRX/IRX-L pairs in Arabidopsis roots. The expression of each

IRX/IRX-L pair was examined using the electronic fluorescenec pictograph (eFP) Browser in
Compare Mode (http://bar.utoronto.ca/efp/cgi-bin/efpWeb.cgi) (Winter et al. 2007) of publicly
available microarray data (Brady et al. 2007). IRX/IRX-L pairs were submitted as either primary
or secondary genes and are colored red or blue, respectively below each root. Tissues colored in
red indicate that the primary gene is more highly expressed, while tissues colored in blue indicate
that the secondary gene is more expressed. The yellow indicates tissues that express the primary
and secondary genes equally. Heat maps are included to show the scale of the log2 ration of the
expression of gene 1 to the expression of gene 2 and color scales associated with these values are
presented
are complex and several TFs cooperate to simultaneously regulate gene expression
of cell wall synthesizing genes. For economical and technical reasons, the first stud-
ies on cell wall TFs targeted secondary cell wall formation. These works identified
a subgroup of plant specific TFs: NAC (NAM/ATAF1/CUC2) and MYB families
that control secondary cell wall thickening (Olsen et al. 2005; Demura and Fukuda
2007; Zhong and Ye 2007). These TFs appear to function as a developmental
switch that induces the transition from primary to secondary cell wall deposition.
VASCULAR-RELATED NAC-DOMAIN1-7 (VND1-7) are members of NAC TF
family that are preferentially expressed in differentiating xylem cells in Arabidopsis
(Kubo et al. 2005), and two of these proteins (VND6 and VND7) act as master
switches for protoxylem and metaxylem development and are negatively regulated
by VND-INTERACTING2 NAC protein (Yamaguchi et al. 2010). No TFs specific
for inducing xylan synthesis have been identified. While five NAC SECONDARY
WALL THICKENING PROMOTING FACTOR 1 (NST1) and NST2, and VND7
proteins appear to function as activators of cell wall genes involved in cellulose,
xylan, and lignin biosynthesis, their direct targets appear to be other transcriptional
activators (Zhong et al. 2008; Ko et al. 2009; McCarthy et al. 2009). Some TFs
such as SND3, SND2, and MYB103 induce the expression of cellulose synthase
genes (Sonbol et al. 2009), but the identification of equivalent TFs for any IRX
genes is still elusive.
Progress is lagging behind in grasses. Currently, only two TFs (ZmMYB31 and
ZmMYB42) are known to affect cell wall biosynthesis. These two maize proteins
have been shown to function as repressors of lignin biosynthesis in Arabidopsis,
but not maize (Fornal et al. 2006; Sonbol et al. 2009). Ambavaram et al. (2011)
showed that the expression of an Arabidopsis TF SHINE gene (a member of the
AP2/ERF family) in rice caused coordinated down-regulation of lignin bio-
synthesis and up-regulation of cellulose and other cell wall biosynthetic genes.
Interestingly, SHINE regulates cutin polymer biosynthesis in Arabidopsis (not
cellulose or lignin) (Kannangara et al. 2007). This underscores the need for con-
tinued efforts to identify and characterize regulatory proteins for GTs associated
with primary and secondary cell wall biosynthesis in various tissues and species.
An attractive strategy for the identification of TFs specific to xylan biosynthesis
in primary cell walls is to use the endosperm tissues in wheat as a model system.
The cereal endosperm represents an excellent model for investigating plant cell
wall biosynthesis and regulation since the walls of cereal endosperm have a sim-
ple composition, consisting mainly of AX, mixed linkage (13),(14)--d-
glucan (MLG), and low amount of cellulose (Philippe et al. 2006a, b; Wilson et
al. 2006). Thus, transcriptomics and/or proteomics of these tissues might allow for
the identification of putative activators and repressors that are specific to AX and/
or MLG synthesis. Several genetic studies have already linked AX biosynthesis
and content in wheat grains to a few puantitative trait loci (QTLs) (Martinant et
al. 1998; Charmet et al. 2009). More recently, Nguyen et al. (2011) conducted an
extensive study on AX content in the grains of the Berkut x Krichauff doubled
haploid (DH) population grown at two different environmental conditions using
a linkage map of 528 genetic markers for QTL mapping. According to this study,
QTLs associated with grain AX contents are located on chromosomes 1A, 2A, 7A,
3D, 4D, and 6B, but carrying only the two QTLs 2A, and 4D was sufficient to sig-
nificantly increase AX content in the grain (Nguyen et al. 2011). Effects of these
two QTLs were further confirmed and validated on other wheat varieties. Two
important QTLs associated with AX accumulation showed epistatic interactions,
which may suggest regulation at the genetic level, as one QTL may contain one
174 A. Faik et al.
or several genes coding for one or several proteins that prevent transcription of the
gene(s) from the other QTL. This would be a major contribution toward under-
standing xylan biosynthesis in grasses.
10.4.2ProteinProtein Interactions: Regulators

of XSCs Assemby
Genetic analyses in Arabidopsis have shown that xylan synthesis can greatly impact
the growth and reproduction of plants at various developmental stages. Thus, it is
not surprising that plants would develop a regulatory step during the assembly and
trafficking of core XSCs in the endomembranes. The importance of IRX and
IRX-L protein interactions (core components of XSCs) can be deduced from
a recent genetic study in Arabidopsis that showed that the truncated Arabidopsis
IRX10 protein (lacking 22 amino acids from its N-terminal) cannot rescue
Arabidopsis irx10/irx10-L double mutants (affected in xylan backbone elongation),
and that the first 32 amino acids of IRX10 protein sequence were necessary for
successful complementation (Wu et al. 2009). Furthermore, two tobacco proteins
NpGUT1 (BAC20928.1) and NtGUT1 (BAD04923.1), homologous to Arabidopsis
IRX10 and IRX10-L, exist in public databases and are both truncated at the
N-terminus (lacking the first 7174 amino acids) (Iwai et al. 2002). The truncated
NpGUT1 could not complement the Arabidopsis irx10/irx10-L double mutant,
but did complement the mutant when fused with the 32-amino acid N-terminus
of IRX10 (Wu et al. 2009). Interestingly, Arabidopsis IRX10 and IRX10-L pro-
teins and their putative orthologs from other plants, are predicted to have a cleav-
able signal peptide, suggesting that these proteins would be soluble in the Golgi
lumen. Therefore, they would require interactions with a Golgi-localized protein for
a proper Golgi targeting. We predict that the sequence between the cleavage site and
the 70th amino acid position play an important role in proteinprotein interactions
within the XSC. Although this hypothesis is confirmed by genetic analysis, it will
need to be confirmed by direct biochemical evidence of such protein interactions.
The general hypothesis is that XSCs assemble in the endoplasmic reticulum
(ER), the location of the synthesis of all secretory proteins. One might expect that
deficiencies in the interactions of XSC proteins would result in the accumula-
tion of some of these proteins in the ER, which in turn may trigger proteasome-
assisted degradation of these proteins. One also might expect that for proper ER
export and Golgi accumulation, the assembled XSCs would need both ER exit and
Golgi retention signals on at least one of the protein complex members. Nothing
is known about xylan synthase trafficking, but the recent discovery of wheat XSC
will allow us to explore these issues. Regulatory mechanisms that maintain an
equilibrium between complex assembly and trafficking would be advantageous for
situations that require a rapid physiological response. For example, controlling the
production of only one key protein of the complex may be sufficient to trigger the
physiological response. Furthermore, nothing is known about the composition of

each XSC (i.e., number of each protein within the complex), or what induces a
proper assembly of a functional XSC. Cell biological tools, such as bimolecular
fluorescence complementation (BiFC) or surface plasmon resonance (SPR), will
be of great help in elucidating the proteinprotein interactions important for regu-
lating XSC trafficking and for XSC complex assembly.
10.4.3Modulation of Xylan Synthesis by Phytohormones:

The Missing Link
Plants use phytohormones to translate environmental changes (external factors)

into physiological responses to adapt to such changes. For example, coleoptiles
and hypocotyls respond to light by inhibiting mesocotyl/hypocotyl cell elongation,
and this light signal is perceived almost exclusively through phytochromes (Pjon
and Furuya 1967; Takano et al. 2001). Interestingly, the inhibition of coleoptile
elongation seems to be a direct result of the inhibition of cell wall polysaccharides
synthesis (Ueda et al. 1994, 1995; Miyamato et al. 1997). Coleoptiles represent an
excellent system to study the regulatory mechanisms of phytohormones on poly-
saccharide synthesis and growth, since coleoptile elongation is under the control
of only three phytohormones: auxin (inducer of cell elongation), abscisic acid
(ABA), and jasmonic acid (JA) (the latter two are both inhibitors of cell elonga-
tion) (Hoffmann-Benning and Kende 1992; Ueda et al. 1995; Giani et al. 1998).
Gibberellins and brassinosteroids do not affect coleoptile development (Toyomasu
et al. 1994; Sekimata et al. 2001). Thus, it will be important to identify the proteins
involved in JA and ABA signaling pathways that lead to the regulation of GAX
biosynthesis in grasses and dicots. The recent characterization of the rice (Oryza
sativa) mutant hebiba, a mutant in JA biosynthesis, provided the first genetic
evidence of crosstalk between light and JA signaling involved in photomorpho-
genesis in monocots (Riemann et al. 2003). In the dark, hebiba seedlings have a
short coleoptile and long mesocotyl, which is opposite the phenotype of wild type
plants (Riemann et al. 2003). Treatment of this mutant with exogenous methyl jas-
monate (Me-JA) restored the wild type phenotype. Furthermore, the Arabidopsis
cev1 mutant has a point mutation in the cellulose synthase A3 (CesSA3) gene, and
showed constitutive production of JA (Ellis and Turner 2001). These findings fur-
ther support the notion that changes in cell wall biosynthesis might be coupled
to the JA-dependent signaling pathway used to induce plant defense responses.
It is tempting to speculate whether JA can modulate (directly or indirectly)
xylan synthesis in grasses, and if so, it would be of great interest to elucidate the
mechanism(s) and the intermediate effectors involved in the process. Although
most JA signal transduction pathways and intermediate effectors have been discov-
ered in Arabidopsis (Thines et al. 2007), we still do not know how these mecha-
nisms and target proteins directly modulate cell wall synthesis in grasses.
176 A. Faik et al.
10.5Concluding Remarks: Impacts of Xylan Biosynthesis

on Biofuel Production
Despite progress in the genetics of xylan synthesis in Arabidopsis, our current

knowledge of the biochemistry of the process is limited. The enzymology, active
sites, functions of the predicted GTs, secretion of the newly synthesized polymer,
coordination of gene expression, and assembly/organization of multi-enzyme com-
plexes for xylan biosynthesis are not well understood. This chapter has discussed
some of the critical aspects of xylan synthesis integrating the current biochemi-
cal and genetic data. Xylans are structurally complex and genetic studies have
demonstrated the importance of these polymers for plant growth and survival.
Manipulating their composition will impact the manner by which they interact
with other polymers within the cell wall, which in turn will strongly influence
energy recovery from plant biomass.
How might manipulating xylan biosynthesis impact biofuel production? We
already know that biofuel yields from grass biomass depend on structural vari-
ations of the chemical composition of xylans. For example, the substitution
rate by GlcA, Ara, or acetylation can either limit the accessibility of hydrolases
to the xylan backbone, or limit fermentable yields by inhibiting yeast fermenta-
tion. Since cellulose microfibrils are thought to be tethered by crosslinking gly-
cans such as xylans, manipulating xylan biosynthesis to produce polymers with
reduced side chain contents should improve biofuel production by making cel-
lulose more accessible. Mortimer et al. (2010) demonstrated that Arabidopsis
gux mutants could produce xylans with very low GlcA branches without affect-
ing plant growth. Although this progress shows potential for simplification of
lignocellulosic biomass conversion in dicots, it still needs to be demonstrated in
grasses. Similarly, many hemicelluloses (including xylans) in dicot cell walls are
acetylated (Goncalves et al. 2008; Scheller and Ulvskov 2010; Lee et al. 2011).
It has been shown that hemicellulose acetylation (including xylans) affects the
downstream sugar conversion process and requires pretreatment of biomass for
biofuel production (Helle et al. 2003; Carroll and Somerville 2009; Selig et al.
2009). Using techno-economic models (Klein-Marcuschamer et al. 2010), Gille
et al. (2011) estimate that a 20% reduction in O-acetylation would lead to a 10%
reduction in the costs of ethanol production.
In grasses, manipulating -(1,3)-arabinosylation of xylans showed the poten-
tial to produce xylans with less Ara side chains but this did not seem to improve
cell wall saccharification in the transgenic wheat plants (Anders et al. 2012). The
only report of improved saccharification through AX manipulation was described
in rice using the xylosyl arabinosyl substitution of xylan (xax1) mutant (Chiniquy
et al. 2012). The XAX1 enzyme is a member of the GT 61 family and catalyzes
the incorporation of -(1,2)-linked Xyl onto Ara side chains to produce Xyl-Ara
side chains. This -(1,2)Xyl usually bears a ferulic acid in grass xylans (Hoije
et al. 2006). Thus, xylans from this rice mutant have lower ferulic acid and Xyl
contents. The lower ferulate content affects xylanxylan and xylan-lignin
cross-linking, which in turn improves saccharification in xax1 mutant plant

(Chiniquy et al. 2012). Unfortunately, the xax1 mutant plants are dwarfed, which
translates to a reduction in total biomass. Therefore, more research is needed to
design strategies to manipulate GAX biosynthesis in grasses for improved digest-
ibility without affecting plant growth and development.
Acknowledgments This material is based upon work partially supported by the National
Science Foundation under Grant No. 1145887 to AF.
References
Ambavaram MMR, Krishnan A, Trijatmiko KR, Pereira A (2011) Coordinated activation of cel-
lulose and repression of lignin biosynthesis pathways in rice. Plant Physiol 155:916931
Andersson SI, Samuelson O (1983) Structure of the reducing end-groups in spruce xylan.
Carbohydr Res 111:283288
Anders N, Wilkinson MD, Lovegrove A, Freeman J, Tryfona T, Pellny TK, Weimar T, Mortimer
JC, Stott K, Baker JM, Defoin-Platel M, Shewry PR, Dupree P, Mitchell RA (2012) Glycosyl
transferases in family 61 mediate arabinofuranosyl transfer onto xylan in grasses. Proc Natl
Acad Sci USA 109:989993
Brady SM, Otlando DA, Lee J-Y, Wang JY, Koch J, Dinneny JR, Mace D, Ohler U, Benefey
PN (2007) A high-resolution root spatiotemporal map reveals dominant expression patterns.
Science 318:801806
Brown DM, Zeef LAH, Ellis J, Goodacre R, Turner SR (2005) Identification of novel genes in
Arabidopsis involved in secondary cell wall formation using expression profiling and reverse
genetics. Plant Cell 17:22812295
Brown DM, Goubet F, Vicky WWA, Goodacre R, Stephens E, Dupree P, Turner SR (2007)
Comparison of five xylan synthase mutants reveals new insight into the mechanisms of xylan
synthesis. Plant J 52:11541168
Brown DM, Zhang Z, Stephens E, Dupree P, Turner SR (2009) Characterization of IRX10 and
IRX10-like reveals an essential role in glucuronoxylan biosynthesis in Arabidopsis. Plant J
57:732746
Brown D, Wightman R, Zhang Z, Gomez LD, Atanassov I, Bukowski J-P, Tryfona T, McQueen-
Mason SJ, Dupree P, Turner SR (2011) Arabidopsis genes IRREGULAR XYLEM (IRX15)
IRX15L encode DUF579-containing proteins that are essential for normal xylan deposition
in the secondary cell wall. Plant J 66:401413
Carpita NC (1996) Structure and biogenesis of the cell walls of grasses. Annu Rev Plant Physiol
Plant Mol Biol 47:445476
Carroll A, Somerville C (2009) Cellulosic biofuels. Annu Rev Plant Biol 60:165182
Charmet G, Masood-Quraishi U, Ravel C, Romeuf I, Balfourier F, Perretant M, Joseph J,
Rakszegi M, Guillon F, Sado P, Bedo Z, Saulnier L (2009) Genetics of dietary fiber in bread
wheat. Euphytica 170:155168
Chiniquy D, Sharma V, Schultink A, Baidoo E, Rautengarten C, Cheng K, Carroll A, Ulvskov P,
Harholt J, Keasling J, Pauly M, Scheller H, Ronald P (2012) XAX1 from glycosyltransferase
family 61 mediates xylosyltransfer to rice xylan. Proc Natl Acad Sci USA 109:1711717122
Costanza M, Baryshnkova A, Bellay J et al (2010) The genetic landscape of a cell. Science
327:425431
Darvill JE, McNeil M, Darvill AG, Albersheim P (1980) Structure of plant cell walls XI.
Glucuronoarabinoxylan, a second hemicellulose in the primary cell walls of suspension-
cultured sycamore cells. Plant Physiol 66:11351139
178 A. Faik et al.
Demura T, Fukuda H (2007) Transcriptional regulation in wood formation. Trends Plant Sci
12:6470
Ebringerova A, Hromadkova Z, Heinze T (2005) Hemicellulose. Adv Polym Sci 186:167
Ellis C, Turner JG (2001) The Arabidopsis mutant cev1 has constitutively active jasmonate and
ethylene signal pathways and enhanced resistance to pathogens. Plant Cell 13:10251033
Faik A (2013) Plant cell wall structure-pretreatment the critical relationship in biomass conver-
sion to fermentable sugars. In: Gu T (ed) Green biomass pretreatment for biofuels produc-
tion. Springer, Berlin, New York, pp 130
Fincher GB (2009) Revolutionary times in our understanding of cell wall biosynthesis and
remodeling in the grasses. Plant Physiol 149:2737
Fischer MH, Yu N, Gray GR, Ralph J, Anderson L, Marlett JA (2004) The gel-forming polysac-
charide of psyllium husk (Plantago ovata Forski). Carbohydr Res 239:20092017
Fornal S, Sonbol F-M, Maes T, Capellades M, Uigdomnech P, Rigau J, Caparrsruiz D (2006)
Down-regulation of the maize and Arabidopsis thaliana caffeic acid O-methyl-transferase
genes by two new maize R2R3-MYB transcription factors. Plant Mol Biol 62:809823
Giani S, Qin X, Faono F, Breviario D (1998) In rice, oryzalin and abscisic acid differentially
affect tubulin mRNA and protein level. Planta 205:334341
Gille S, de Souza A, Xiong G, Benz M, Cheng K, Schultink A, Reca I-B, Pauly M (2011)
O-Acetylation of Arabidopsis hemicellulose xylogucan requires AXY4 or AXY4L proteins
with a TBL and DUF231 domain. Plant Cell 23:40414053
Gille S, Pauly M (2012) O-acetylation of plant cell wall polysaccharides. Front Plant Sci 3:12.
doi:10.3389/fpls.2012.00012
Goncalves VMF, Evtuguin DV, Rosario M, Domingues M (2008) Structural characterization of
the acetylated heteroxylan from the natural hybrid Paulownia elongata/Paulownia fortunei.
Carbohydr Res 343:256266
Goubet F, Jackson P, Deery MJ, Dupree P (2002) Polysaccharide analysis using carbohydrate gel
electrophoresis: a method to study plant cell wall polysaccharides and polysaccharide hydro-
lases. Anal Biochem 300:5368
Helle S, Cameron D, Lam J, White B, Duff S (2003) Effect of inhibitory compounds found in
biomass hydrolysates on growth and xylose fermentation by a genetically engineered strain
of S. cerevisiae. Enzyme Microb Technol 33:786792
Higa HH, Butor C, Diaz S, Varki A (1989) O-acetylation and de-O-acetylation of sialic acids.
O-acetylation of sialic acids in the rat liver Golgi apparatus involves an acetyl intermedi-
ate and essential histidine and lysine residues- a transmembrane reaction? J Biol Chem
264:1942719434
Hoffmann-Benning S, Kende H (1992) On the role of abscisic acid and gibberellin in the regula-
tion of growth in rice. Plant Physiol 99:11561161
Hoije A, Sandstrom C, Roubroeks J, Andersson R, Gohil S, Gatenholm P (2006) Evidence of
the presence of 2-O--d-xylopyranose--l-arabinofuranose side chains in barley husk arabi-
noxylan. Carbohydr Res 341:29592966
Iwai H, Masaoka N, Ishii T, Satoh S (2002) A pectin glucuronyltransferase gene is essential for
intracellular attachment in the plant meristem. Proc Natl Acad Sci USA 99:1631916324
Jensen JK, Kim H, Cocuron J-C, Orler R, Ralph J, Wilkerson CG (2011) The DUF579 domain
containing proteins IRX15 and IRX15-L affect xylan synthesis in Arabidopsis. Plant J
66:387400
Johansson MH, Samuelson O (1977) Reducing end groups in birch xylan and their alkaline deg-
radation. Wood Sci Technol 11:261263
Joseleau J-P, Comtat J, Ruel K (1992) Chemical structure of xylans and their interaction in the
plant cell walls. In: Visser J et al (eds) Xylans and xylanases. Elsevier, Amsterdam, pp 115
Joshi CP, Mansfield SD (2007) The cellulose paradox-simple molecule, complex biosynthesis.
Curr Opin Plant Biol 10:220226
Jones L, Ennos AR, Turner SR (2001) Cloning and characterization of irregular xylem4 (irx4): a
severely lignin-deficient mutant of Arabidopsis. Plant J 26:205216
Kannangara R, Branigan C, Liu Y, Penfield T, Rao V, Mouille G, Hofte H, Pauly M, Riehmann

JL, Broun P (2007) The transcription factor WIN1/SHN1 regulates cutin biosynthesis in
Arabidopsis thaliana. Plant Cell 19:12781294
Kiefer L, York W, Darvill A, Albersheim P (1989) Xyloglucan isolated from suspension-cultured
sycamore cell walls is O-acetylated. Phytochemistry 28:21052107
Klein-Marcuschamer D, Oleskowicz-Popiel P, Simmons BA, Blanch HW (2010)
Technoeconomic analysis of biofuels: a wiki-based platform for lignocellulosic biorefineries.
Biomass Bioenergy 34:19141921
Ko J-H, Kim W-C, Han K-H (2009) Ectopic expression of MYB46 identifies transcriptional reg-
ulatory genes involved in secondary wall biosynthesis in Arabidopsis. Plant J 60:649665
Kubo M, Udagawa M, Nishikubo N, Horiguchi G, Yamaguchi M, Ito J, Mimura T, Fukuda H,
Demura T (2005) Transcription switches for protoxylem and metaxylem vessel formation.
Genes Dev 19:18551860
Lee C, Zhong R, Richardson EA, Himmelsbach D, Mcphail BT, Ye ZH (2007) The PARVUS
gene is expressed in cells undergoing secondary wall thickening and is essential for glucu-
ronoxylan biosynthesis. Plant Cell Physiol 48:16591672
Lee C, Teng Q, Huang W, Zhong R, Ye ZH (2009) The F8H glycosyltransferase is a functional
paralog of FRA8 involved in glucuronoxylan biosynthesis in Arabidopsis. Plant Cell Physiol
50:812827
Lee C, Teng Q, Zhong R, Ye ZH (2011) The four Arabidopsis REDUCED WALL
ACETYLATION genes are expressed in secondary wall-containing cells and required for
acetylation of xylan. Plant Cell Physiol 52:12891301
Martinant JP, Cadalen T, Billot A, Chartier S, Leroy P, Bernard M, Saulnier L, Brankard A
(1998) Genetic analysis of water-extractable arabinoxylans in bread wheat endosperm. Theor
Appl Genet 97:10691075
McCarthy RL, Zhong R, Ye Z-H (2009) MYB83 is a direct target of SND1 and acts redundantly
with MYB46 in the regulation of secondary cell wall biosynthesis in Arabidopsis. Plant Cell
Physiol 50:19501964
Miyamoto K, Oka M, Ueda J (1997) Update on the possible mode of action of the jasmonates:
focus on the metabolism of cell wall polysaccharides in relation to growth and development.
Physiol Plant 100:631638
Mortimer JC, Miles GP, Brown DM, Zhang Z, Segura MP, Weimar T, Yu X, Seffen KA, Stephens
E, Turner SR, Dupree P (2010) Absence of branches from xylan in Arabidopsis gux mutants
reveals potential for simplification of lignocellulosic biomass. Proc Natl Acad Sci USA
40:1740917414
Nguyen V-L, Huynh B-L, Wallwork H, Stangoulis J (2011) Identification of quantitative trait loci
for grain arabinoxylan concentration in bread wheat. Crop Sci 51:11431150
Olsen AN, Ernst HA, Leggio LL, Skriver K (2005) NAC transcription factors: structurally dis-
tinct, functionally diverse. Trends Plant Sci 10:7987
Pena MJ, Zhong R, Zhou G-K, Richardson EA, ONeill MA, Darvill AG, York WS, Ye Z-H
(2007) Arabidopsis irregular xylem8 and irregular xylem9: Implications for the complexity
of glucuronoxylan biosynthesis. Plant Cell 19:549563
Persson S, Caffall KH, Freshour G, Hilley MT, Bauer S, Poindexter P, Hahn MG, Mohnen D,
Somerville C (2007) The Arabidopsis irregular xylem8 mutant is deficient in glucuronoxy-
lan and homogalacturonan, which are essential for secondary cell wall integrity. Plant Cell
19:237255
Philippe S, Saulnier L, Guillon F (2006a) Arabinoxylan and (1,3)(1,4)--glucan deposition in cell
walls during wheat endosperm development. Planta 224:449461
Philippe S, Tranquet S, Utille J-P, Saulnier L, Guillon F (2006b) Investigation of ferulate deposi-
tion in endosperm cell walls of mature and developing wheat grains using a polyclonal anti-
body. Planta 225:12871299
Pjon C-J, Furuya M (1967) Phytochrome function in Oryza sativa L. I. Growth responses of etio-
lated coleoptiles to red, far-red and blue light. Plant Cell Physiol 8:709718
180 A. Faik et al.
Riemann M, Muller A, Korte A, Furuya M, Weilwer EW, Nick P (2003) Impaired induction of
the jasmonate pathway in the rice mutant hebiba. Plant Physiol 133:18201830
Selig M, Adney WS, Himmel M, Decker S (2009) The impact of cell wall acetylation on corn
stover hydrolysis by cellulolytic and xylanolytic enzymes. Cellulose 16:711722
Sekimata K, Kimura T, Kaneko I, Nakano T, Yoneyama K, Takeuchi Y, Yoshida S, Asami T
(2001) A specific brassinosteroid biosynthesis inhibitor, Brz 2001: evaluation of its effects on
Arabidopsis, cress, tobacco, and rice. Planta 213:716721
Scheller HV, Ulvskov P (2010) Hemicelluloses. Annu Rev Plant Biol 61:263289
Somerville CR (2006) Cellulose synthesis in higher plants. Ann Rev Cell Dev Biol 22:5378
Sonbol F-M, Fornal S, Capellades M, Encina A, Tourio S, Torres J-L, Rovira P, Ruel K,
Puigdomnech P, Rigau J et al (2009) The maize ZmMYB42 represses the phenylpropanoid
pathway and affects the cell wall structure, composition and degradability in Arabidopsis
thaliana. Plant Mol Bio 70:283296
Takano M, Kanegae H, Shinomura T, Miyao A, Hirochika H, Furuya M (2001) Isolation and
characterization of rice phytochrome A mutants. Plant Cell 13:521534
Thines B, Katser L, Melotto M, Niu Y, Mandaokar A, Liu G, Nomura K, He S-Y, Howe GA,
Browse J (2007) JAZ repressor proteins are targets of the SCFCOI1 complex during jasmonate
signaling. Nature 448:661666
Toyomasu T, Yamane H, Murofushi N, Nick P (1994) Phytochrome inhibits the effectivness of
gibberellins to induce cell elongation in rice. Planta 194:256263
Turner SR, Somerville CR (1997) Collapsed xylem phenotype of Arabidopsis identifies mutants
deficient in cellulose deposition in the secondary cell wall. Plant Cell 9:689701
Ueda J, Miyamoto K, Aoki M (1994) Jasmonic acid inhibits the IAA-induced elongation of oat
coleoptile segments: a possible mechanism involving the metabolism of cell wall polysaccha-
rides. Plant Cell Physiol 35:10651070
Ueda J, Miyamoto K, Kamisaka S (1995) Inhibition of the synthesis of cell wall polysaccha-
rides in oat coleoptile segments by jasmonic acid: relevance to its growth inhibition. J Plant
Growth Reg 14:6976
Urbanowicz BR, Pena MJ, Ratnaparkhe S, Avci U, Backe J, Steet HF, Foston M, Li H, ONeill
MA, Ragauskas AJ, Darvill AG, Wyman C, Gilbert HJ, York WS (2012) 4-O-methylation of
glucuronic acid in Arabidopsis glucuronoxylan is catalyzed by a domain of unknown func-
tion family 579 protein. Proc Natl Acad Sci USA 109:1425314258
Wilson SM, Burton RA, Doblin MS, Stones BA, Newbigin EJ, Fincher GB, Bacic A (2006)
Temporal and spatial appearance of wall polysaccharides during cellularization of barley
(Hordeum vulgare) endosperm. Planta 224:655667
Winter D, Vinegar B, Nahal H, Ammar R, Greg V, Wilson GV, Provart NJ (2007) An electronic
fluorescent pictograph browser for exploring and analyzing large-scale biological data sets.
PLoS ONE 2:e718. doi:10.1371/journal.pone.0000718
Wu AM, Rihouey C, Seveno M, Hornblad E, Singh SK, Matsunaga T, Ishii T, Lerouge P,
Marchant A (2009) The Arabidopsis IRX10 and IRX10-like glycosyltransferases are critical
for glucuronoxylan biosynthesis during secondary cell wall formation. Plant J 57:718731
Wu AM, Hornblad E, Voxeur A, Gerber L, Rihouey C, Lerouge P, Marchant A (2010) Analysis
of the Arabidopsis IRX9/IRX9-L and IRX14/IRX14-L pairs of glycosyltransferase genes
reveals critical contributions to biosynthesis of the hemicellulose glucuronoxylan. Plant
Physiol 153:542554
Yamaguchi M, Ohtani M, Mitsuda N, Kubo M, Ohme-Takagi M, Fukuda H, Demura T (2010)
VND-INTERACTING2, a NAC domain transcription factor, negatively regulates xylem ves-
sel formation in Arabidopsis. Plant Cell 22:12491263
Zablackis E, Huang J, Muller B, Darvill AG, Albersheim P (1995) Structure of plant-cell walls:
characterization of the cell-wall polysaccharides of Arabidopsis thaliana leaves. Plant
Physiol 107:11291138
Zeng W, Chatterjee M, Faik A (2008) UDP-xylose-stimulated glucuronyltransferase activity in
wheat microsomal membranes: characterization and role in glucurono(arabino)xylan biosyn-
thesis. Plant Physiol 147:7891
Zeng W, Jiang N, Nadella R, Killen TL, Nadella V, Faik A (2010) A glucurono(arabino)xylan

synthase complex from wheat (Triticum aestivum L.) contains members of the GT43, 47, and
75 families and functions cooperatively. Plant Physiol 154:7897
Zhong R, Ye Z-H (2007) Regulation of cell wall biosynthesis. Curr Opin Plant Biol 10:564572
Zhong R, Lee C, Zhou J, McCarthy RL, Ye Z-H (2008) A battery of transcription factors
involved in the regulation of secondary cell wall biosynthesis in Arabidopsis. Plant Cell
20:27632782
Chapter 11
Towards Redesigning Cellulose Biosynthesis
for Improved Bioenergy Feedstocks
Catherine Rayon, Anna T. Olek and Nicholas C. Carpita
Abstract With an estimated 200 billion tons produced annually, cellulose is the
most abundant biopolymer on earth. Cellulose is expected to be the principal feed-
stock for liquid biofuels and bio-based products, but its para-crystalline nature
results in recalcitrance to deconstruction required for biological and chemical
conversion to useful products. Recent work solving the 3D structure of a bacte-
rial cellulose synthase, modeling of plant cellulose synthases, and the 3D contour
structure of the catalytic domain of a plant cellulose synthase have contributed
new perspectives on the organization of catalytic units in the rosette complex.
These discoveries stimulate new approaches to engineer the complex to make
altered forms of cellulose for enhancing efficiency of biomass deconstruction for
biofuel production or for synthesis of new materials and nanoproducts.
Keywords Lignocellulosic biomass Cellulose recalcitrance Cellulose bio-

synthesis Cellulose synthase
Lignocellulosic biomass, the thickened secondary cells of vascular plants, is the

primary source of reduced carbon for biofuels and bio-based products (Sarkar
et al. 2009). Chemically, lignocellulosic cell wall is a complex organization of a
para-crystalline cellulose microfibrils interlaced with cross-linking glycans, typi-
cally xylans in angiosperms and mannans in gymnosperms (Pauly and Keegstra
2008). This polysaccharide framework interacts with a co-extensive matrix of
lignin to form architectural structures that provide fiber strength, resistance to
extreme tensions in tracheary elements created by transpirational pull of water,
C. Rayon
EA 3900-BIOPI, Universit de Picardie Jules Verne, 80039 Amiens, France
A. T. Olek N. C. Carpita(*)
Department of Botany and Plant Pathology, Purdue University, 915 West State Street,
West Lafayette, IN 47907-2054, USA
e-mail: carpita@purdue.edu
184 C. Rayon et al.
and hydrophobic barriers that maintain columns of water required for survival
(Boerjan et al. 2003). A wealth of information about how lignin formation can be
modified comes from studies vascular and fiber cell identity, where timing and
balance of transcription regulation through a cascade of NAC and MYB domain-
containing factors and microRNA expression define the lignin and polysaccharide
composition of secondary walls during vascularization (Brady et al. 2010; Zhong
et al. 2010; Zhao and Dixon 2011). Understanding the fine control of networks
offers many opportunities and strategies to engineer biomass composition more
precisely.
Considerable research effort is devoted to reducing the recalcitrance of cell
walls to conversion to ethanol and advanced biofuels. As described above, much
of this recalcitrance is attributed to lignin interactions that prevent hydrolysis by
digestive enzymes, but the high crystallinity of cellulose and its interactions with
cross-linking glycans and lignin also constitutes additional targets for enhancing
efficiency of deconstruction and conversion (Himmel et al. 2007). Our focus here
is on progress in understanding the mechanism of synthesis and assembly of cel-
lulose to gain the knowledge needed to develop strategies to redesign this foun-
dational structure of plant cell wall architecture to be an optimal raw material for
biofuel production.
11.1The Construction, Delivery and Turnover of Cellulose

Synthase Complexes
Cellulose synthesis begins with the expression of several primary- and secondary
wall-specific isoforms of the CesA and their assembly into complexes at the Golgi
membrane (Haigler and Brown 1986). We have long known that a continuous sup-
ply of these complexes is required to maintain synthesis, indicating that each CSC
might synthesize one or a limited number of microfibrils (Herth 1985; Haigler and
Brown 1986). Demonstration that cellulose synthases have extremely short half-
lives (on the order of 20min) supports this view (Jacob-Wilk et al. 2006). A major
distinction of plant cells is that they move large numbers of vesicle clusters from
Golgi to PM during cell wall synthesis (Toyooka et al. 2009). The alternate path-
way of the CSC trafficking via cortical microtubules to the PM in contrast to actin-
based trafficking of the non-cellulosic polysaccharides offers some opportunities
for control of the relative abundance of each in the developing wall. Cytoskeletal
elements associated with CesAs including kinesins (Zhong et al. 2010) and
dynamin-related proteins (Collings et al. 2008; Hirano et al. 2010; Taylor 2011).
We are just learning how CSC internalization via dynamins is associated with a
clathrin-dependent pathway in support of the turnover (Fujimoto et al. 2010).
From in vivo labeling and in vitro biosynthesis dating back to the 1960s, we
have known that primary and secondary wall non-cellulosic non-cellulosic
cross-linking glycans and pectins are synthesized at the Golgi and transported in
large secretory vesicles via actin filaments to the plasma membrane where they
11 Towards Redesigning Cellulose Biosynthesis 185
are assembled around newly synthesized cellulosic microfibrils (for review, see
Carpita 2011). More recently we have learned that cellulose synthase complexes
are packaged and exported by a different route involving small vesicles and corti-
cal microtubules (Gutierrez et al. 2009; Crowell et al. 2009). Live-cell imaging
shows a cellulose synthase interactive protein (CSI1) co-localizes with CesAs
and moves bidirectionally with them (Gu et al. 2010). CSI1 is a microtubule-
associated protein that forms a direct bridge to the cellulose synthase complex,
and disruption of this connection in the csi1 mutant results in dissociation of co-
alignments of CSCs and the microtubule, suggesting both CSC delivery and guid-
ance functions for microfibril deposition (Li et al. 2012). Both the N-terminal
Zn-finger and catalytic domains of CesA have potential phosphorylation sites at
Ser residues, and mutation of these sites to mimic and a permanent on or off state
results in severe alterations in cell elongation and velocities of tracking of fluores-
cence-tagged CesAs (Chen et al. 2010). In summary, discovery of the dynamics of
cellulose synthases, their special routes of its trafficking to the plasma membrane,
and their activation mechanisms through phosphorylation each provide targets for
manipulation to enhance rates of synthesis upon delivery to the plasma membrane.
However, they do not in themselves provide a means of manipulation of cellu-
lose microfibril structure. For that we need to learn more about the mechanism of
synthesis.
11.2The Cellulose Synthesis at the Plasma Membrane
To redesign the cell wall as an optimal feedstock requires a more complete under-
standing of the biology of the cellulose synthase complex, from the biochemical
mechanism of the polymerization of the (14)--D-glucan chains, to the coor-
dination of the crystallization process, and to the higher order bundling that occurs
in muro. In plants, individual -glucan chains of cellulose are synthesized and
crystallized at the plasma membrane by large membrane complexes termed parti-
cle rosettes (Giddings et al. 1980; Mueller and Brown 1980). Particle rosettes are
six-membered hexagonal arrays about 25nm in diameter and represents the mem-
brane spanning domains and short extracellular loops of roughly three dozen cel-
lulose synthase polypeptides (Fig.11.1). Cellulose synthases (CesAs) are intrinsic
membrane proteins with their catalytic domains extending into the cytoplasm.
Membrane foot-printing gives estimated diameters of about 50nm for the col-
lection of 36 cellulose synthase polypeptides (Bowling and Brown 2008). The 36
-glucan chains from the rosette complex, i.e. six chains per particle, give an esti-
mated microfibril diameter of around 3.63.8nm for most primary wall cellulose
(Kennedy et al. 2007). More recent evidence employed solid-state 13C-nuclear
magnetic resonance (NMR) spectroscopy and neutron and x-ray diffraction indi-
cate a smaller microfibril diameters on the order of 2.73.0nm, which indicate
that the crystalline portions of the microfibril are only 1824 glucans (Fernandes
et al. 2011; Thomas et al. 2013). As these techniques are relevant for the
186 C. Rayon et al.
Fig.11.1Particle rosette structures of plant cellulose synthase complexes. Freeze-etch images

of the P-face of the plasma membrane showing clusters of rosettes associated with the develop-
ing of secondary wall spiral thickenings of a Lepidium tracheary element (from Herth 1985).
The inset shows the 6-fold symmetry of a single particle rosette from a Zinnia tracheary element
developing in vitro (from C. Haigler, unpublished data, as seen in Delmer 1999). A substructure
can be observed in each of the particles. In these freeze-etch images, only the membrane-span-
ning domains and extracellular loops of the CesA proteins can be observed
crystalline domains, the inference is that as many as 18 of the glucan chains sur-
face coat the microfibril in a non-crystalline, flexible way that is capable of inter-
acting tightly with non-cellulosic glycans, such as xyloglucan, glucomannan and
xylan, or other microfibrils to form bundles. Bundling of both primary and second-
ary wall microfibrils can be extensive, creating a higher order of complexity and
recalcitrance to deconstruction (Ding and Himmel 2006). Targets of opportunity
for altering the microfibril structure include introduction of alternative sugars or
linkages into the microfibril during synthesis to disrupt the para-crystalline array,
altering the number of -glucan chains per microfibril to minimize diameters, and
interference with bundling with altered non-cellulosic glycans that alter nano-scale
architecture without radically altering the functional structure of the cell wall.
11.3The Mechanism of Cellulose Synthesis
Particle rosettes comprise several dozen cellulose synthase polypeptides of about

110kDa, each with a large, cytoplasmic N-terminal region containing a zinc-
finger (ZnF) domain involved in coupling CesAs (Kurek et al. 2002), and eight
Fig.11.2The CesA gene family of Arabidopsis. a Domain model for a CesA. Two ZnF
domains (in yellow) are found in the N terminus before the first membrane-spanning domain (in
blue). The large central domain contains four highly conserved catalytic motifs (formerly called
U motifs) of D, DxD, (TE) D, and QxxRW, important for substrate binding and catalysis. The
class-specific region (CSR) is conserved among orthologs of the same subclade and vary in the
number of upstream conserved Cys residues, the number of consecutive basic amino acids, Lys
and Arg, and the number of consecutive acidic amino acids, Asp and Glu, downstream from the
basic residues (after Carpita and Vergara 1998; Vergara and Carpita 2001). Eight transmembrane
domains, two upstream and six downstream of the catalytic domain, are predicted to interact to
form a channel through which a single -glucan chains are secreted to the cell wall. However, the
finding that catalytic dimers form the fundamental unit of construction suggests that the mem-
brane-spanning domains might fuse into a larger channel to extrude two chains instead of one. b
Orthologous relationships of CesAs of Arabidopsis (red), rice (green), and maize (blue). Of the
ten Arabidopsis CesA genes, at least three are coexpressed during primary wall formation, and
mutations in each of them, AtCesA1 (RSW1; At4g32410), AtCesA6 (PRC1; At5g64740), and
AtCesA3 (CEV1 and ELI1; At5g05170), result in cellulose deficiencies and demonstrate non-
rdundancy. The irx mutants AtCesA8 (IRX1; At4g18780), AtCesA7 (IRX3; At5g17420), and
AtCesA4 (IRX5; At5g44030), which are also non-redundant, are deficient in cellulose synthesis
specifically in secondary walls (after Penning et al. 2009)
membrane spans, two upstream and six downstream, sandwiching a catalytic

domain (CatD) containing four catalytic motifs, formerly called U-motifs, with D,
DxD, D, and QxxRW (Fig.11.2; Delmer 1999). Within the catalytic domain are
the Plant-Conserved Sequence (P-CR), located between the first D and DxD motif,
and the Class-Specific Region (CSR), for which high similarity across many spe-
cies was observed among subclasses of orthologous isoforms (Fig.11.2b; Vergara
and Carpita 2001).
188 C. Rayon et al.
Plant CesAs share homology with sequences of bacterial CesA proteins with
respect to the four catalytic motifs essential for substrate binding and catalysis
(Saxena et al. 1995; Pear et al. 1996). The biochemical mechanism of cellulose syn-
thesis in plants must solve three fundamental physical problems: (a) synthesis of
the (14)--D-glucosyl linkage requires that each glucosyl residue is turned 180
with respect to its neighboring sugar, (b) a membrane channel of sufficient size is
needed to permit extrusion of the glucan chain, and (c) a mechanism is needed to
couple the many synthases into the rosette complex (Carpita 2011). The structure
of the Rhodobacter sphaeroides CesA (BcsA) solves the first two problems with a
single-site mechanism that toggles between two conformations of the non-reducing
end acceptor glucosyl residue, spiraling the chain through the 8-membered mem-
brane channel into a second protein, BcsB, that guides the chain through the peri-
plasmic space (Morgan et al. 2013). However, the BcsA synthase functions as a
monomer, and plants do not contain the accessory protein but extrude the chains
directly to the extracellular surface where they are organized by an unknown mech-
anism into microfibrils. In contrast, plant CesAs couple into complexes via three
additional sequences not present in bacterial CesAs, the Zn-finger domains, the
CSR, and/or P-CR. Recently, we showed that recombinant expression of only the
large catalytic domains between membrane-spanning domains II and III can be iso-
lated as monomers in the presence of thiol reducing agents but dimerize when the
proteins are concentrated or the thiol reducing agent depleted (Olek et al. 2013).
Small-angle x-ray scattering (SAXS) of the monomer shows an elongated two-
domain structure, with dimers coupled through the smaller domains.
The 3D crystal structure of the BcsA synthase gives a significant conformation
of the amino acids that function in UDP-Glc binding, chain termination position-
ing and catalysis of glycosyl transfer (Morgan et al. 2013). Structure modeling of
a cotton secondary wall CesA showed good conservation of the active site defined
by the four catalytic motifs and other amino acids with the BcsA (Sethaphong et
al. 2013). Olek et al. (2013) used several threading and structure prediction mod-
els to show the catalytic motifs give quite similar active site conservation with the
BcsA when a rice CesA is truncated to remove P-CR and CSR domains not pre-
sent in the bacterial protein. However, considerable uncertainty remains concern-
ing the P-CR and CSR domains subtending the active site because of the lack of
specific templates. Ab initio modeling (Sethaphong et al. 2013) versus composite
modeling against chimeric templates (Olek et al. 2013) gave drastically different
models for the conformation of the P-CR and CSR.
Sethaphong et al. (2013) extend their model to form a symmetrical hexamer
that involves coupling of the CSR and P-CR. In contrast, Olek et al. (2013) predict
dimerization through the P-CR and/or CSR domains form the fundamental unit of
synthesis, and three dimers couple by complementary Zn-finger domains to form
a single particle of the 6-membered particle rosette. Interaction of Zn-fingers is
precluded in the symmetrical hexamer. In either model, each monomer synthesizes
a single -glucan chain by a single catalytic mechanism that toggles between the
alternating position of O-4 of the non-reducing terminal glucose needed to form
the (14)--D-linkage. However, based on the contour structure predicted by
SAXS, the fundamental dimer hypothesis predicts the UDP-Glc entry from the
outward faces and extrusion of two chains through the cavity formed by the dimer
(Olek et al. 2013).
While the 3D crystal structure of a plant CesA is needed to determine the roles
of the P-CR and CSR unequivocally, it is reasonable to predict that the mecha-
nism of synthesis and the structure of the active site of the bacterial synthase
is conserved in the plant enzyme. Thus, amino acid targets for manipulation of
the linkage would be those predicted to be involved with toggling mecha-
nism while maintaining the uridinyl binding domain of UDP-Glc (Morgan et al.
2013). Mapping the differences in the mechanism of (13)--D-glucan vs.
(1 4)--D-glucan synthases could provide some guidance for protein engi-
neering. One could contemplate protein engineering strategies to convert the
uridinyl binding domain to a guanidinyl binding domain to accommodate a GDP-
Man. Man is a C-2 epimer of Glc and may need some remodeling of the toggling
domain to accommodate the axial -OH of Man rather than the equatorial -OH of
Glc. Man in place of Glc could potentially introduce slight alterations in the crys-
tal structure of cellulose that would ease deconstruction without modifying the
fundamental microfibril function.
The formation of catalytic dimers as the fundamental units of synthesis gives a
new perspective on the requirement for multiple isoforms of the CesA associated
with primary and secondary wall cellulose formation. As mutants of each impair
synthesis, their association with the synthase complex is non-redundant (Taylor
2003), and direct interactions of three distinct CesA polypeptides have been shown
in vivo by bimolecular fluorescence complementation (Desprez et al. 2007) and
in vitro by affinity pull-down experiments (Taylor et al. 2003; Wang et al. 2008;
Atanassov et al. 2009), and yeast 2-hybrid studies (Timmers et al. 2009). The key
question becomes whether or not certain isoforms form better homo- or heter-
odimers, and the role domain swapping or active-site manipulation might play in
modulating the number of chains produced.
11.4Accessory Proteins that Might Function in Cellulose

Synthesis
Amor et al. (1995) proposed that a sucrose synthase (SuSy) associated with the
plasma membrane constituted a UDP-Glc metabolic channeling mechanism for
cellulose synthase. This idea was strengthened by demonstration immunologically
of SuSy in the rosette structures (Fujii et al. 2010). However, several lines of con-
flicting data have cast doubt on a requirement for SuSy in cellulose synthesis. A
quadruple mutant that eliminates all detectable SuSy has no effect on rates of cel-
lulose synthesis in Arabidopsis (Barratt et al. 2009). Baroja-Ferandz et al. (2012)
challenged this conclusion by claiming that the reaction conditions for SuSy were
not optimized and that even the quadruple mutant had sufficient SuSy activity to
support cellulose synthase. Smith et al. (2012) argued that the activity measured
190 C. Rayon et al.
was from two additional isoforms that are strictly phloem localized and irrelevant
for mesophyll cells of the leaf. Regardless of a facilitative or required role for
SuSy as a metabolic channel for UDP-Glc to cellulose synthase, over-expression
of SuSy in transgenic poplar results in significant increases in cellulose content
(Coleman et al. 2009). Thus, enhancing amounts of SuSy in the absence of over-
expression of the components of the cellulose synthase complex is a reasonable
strategy to enhance yields of cellulose.
Several interactions of CesAs with other proteins have been inferred from
mutants whose phenotypes include disruption of cellulose synthesis. Loss of tri-
chome birefringence (TBR), long-recognized as a cellulose deficiency (Potikha
and Delmer 1995), was traced to a gene encoding a transmembrane domain-
containing protein with a domain of unknown function (DUF231) (Bishoff
et al. 2010). TBR is a member of a very large family that also contains genes that
encode senescence-related, repression of freezing-tolerance, and resistance to
powdery mildew.
KORRIGAN, a transmembrane-containing endo-(14)--D-glucanase
required for cell growth and cellulose deposition, co-localizes with CesAs at the
plasma membrane but remains without a specific function (Crowell et al. 2010).
Members of COBRA, a large gene family encoding glycosyl phosphatidylinosi-
tol (GPI)-anchored membrane associated protein, are necessary for normal cell
development and cell-wall architecture, with different isoforms associated with
normal primary- or secondary-wall cellulose biosynthesis in grasses (Schindelman
et al. 2001; Roudier et al. 2005; Sato et al. 2010). While the co-localize to the
plasma membrane with CesAs, they do not appear to function directly in synthesis
of the glucan chains, but in the orientation and patterning of both cellulose and
lignin during wall deposition. Their absence gives rise to stem brittleness with-
out change in tensile strength in stress-strain experiments (Sindhu et al. 2007).
As some studies indicate that COBRAs might function in crystallization, there
is potential to manipulation to alter recalcitrance properties of cellulose. Thus,
manipulation of secondary wall COBRAs might have a utility in wall densification
or fragmentation during processing.
11.5Conclusions
Good progress has been made in unraveling the intricacies of the protein structure,
synthase assembly, trafficking, and synthase mechanisms of cellulose synthase
complexes. At each step of resolution, researchers are given important insights
on how the features of microfibril assembly could be altered to improve biomass
as a feedstock for biofuels and bio-based products. However, in a broader sense
unraveling the complexities of transcriptional networks that are responsible for
establishing the diverse lignocellulosic compositions and architectures found in
any species holds special promise for redesigning the cell wall for deconstruction
to valued end-products.
Acknowledgments This review was completed through support of the Center for Direct
Catalytic Conversion of Biomass to Biofuels, an Energy Frontier Research Center funded by
the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences (award no.
DESC0000997).
References
Amor Y, Haigler CH, Johnson S, Wainscott M, Delmer DP (1995) A membrane-associated form

of sucrose synthase and its potential role in synthesis of cellulose and callose in plants. Proc
Nat Acad Sci USA 92:93539357
Atanassov II, Pittman JK, Turner SR (2009) Elucidating the mechanisms of assembly and subu-
nit interaction of the cellulose synthase complex of Arabidopsis secondary cell walls. J Biol
Chem 284:38333841
Baroja-Fernndez E, Muoz FJ, Li J, Bahaji A, Almagro G, Montero M, Etxeberria E,
Hidalgo M, Sesma MT, Pozueta-Romero J (2012) Sucrose synthase activity in the
sus1/sus2/sus3/sus4 Arabidopsis mutant is sufficient to support normal cellulose and starch
production. Proc Nat Acad Sci USA 109:321326
Barratt DHP, Derbyshire P, Findlay K, Pike M, Wellner N, Lunn J, Feil R, Simpson C, Maule AJ,
Smith AM (2009) Normal growth of Arabidopsis requires cytosolic invertase but not sucrose
synthase. Proc Nat Acad Sci USA 106:1312413129
Bischoff V, Silvia N, Neumetzler L, Schindelasch D, Urbain A, Eshed R, Persson S, Delmer D,
Scheible WR (2010) TRICHOME BIREFRINGENCE and its homolog At5g01360 encode
plant-specific DUF231 proteins required for cellulose biosynthesis in Arabidopsis. Plant
Physiol 153:590602
Boerjan W, Ralph J, Baucher M (2003) Lignin biosynthesis. Annu Rev Plant Biol 54:519546
Bowling AJ, Brown RM (2008) Cytoplasmic domain of cellulose synthesizing complexes in
plants. Protoplasma 233:115127
Brady SM, Zhang L, Megraw M, Martinez NJ, Jiang E, Yi CS, Liu W, Zeng A, Taylor-Teeples
M, Kim D et al (2010) A stele-enriched gene regulatory network in the Arabidopsis root. Mol
Syst Biol 7: Art. No. 459
Carpita NC, Vergara CE (1998) A recipe for cellulose. Science 279:672673
Carpita NC (2011) Update on mechanisms of plant cell wall biosynthesis: how plants make cel-
lulose and other (14)--D-glycans. Plant Physiol 155:171184
Chen S, Ehrhardt DW, Somerville CR (2010) Mutations of cellulose synthase (CESA1) phos-
phorylation sites module anisotropic cell expansion and bidirectional mobility of cellulose
synthase. Proc Nat Acad Sci USA 107:1718817193
Coleman HD, Yan J, Mansfield SD (2009) Sucrose synthase affects carbon partitioning to increase
cellulose production and altered cell wall ultrastructure. Proc Nat Acad Sci USA 106:13118
Collings DA, Gebbie LK, Howles PA, Hurley UA, Birch RJ, Cork AH, Hocart CH, Arioli T,
Williamson RE (2008) Arabidopsis dynamin-like protein DRP1A: a null mutant with wide-
spread defects in endocytosis, cellulose synthesis, cytokinesis, and cell expansion. J Exp Bot
59:361376
Crowell EF, Bischoff V, Desprez T, Rolland A, Stierhof Y-D, Schumacher K, Gonneau M, Hfte
H, Vernhettes S (2009) Pausing of Golgi bodies on microtubules regulates secretion of cellu-
lose synthase complexes in Arabidopsis. Plant Cell 21:11411154
Crowell EF, Gonneau M, Stierhof Y-D, Hfte H, Vernhettes S (2010) Regulated trafficking of cel-
lulose synthases. Curr Opin Plant Biol 13:700705
Delmer DP (1999) Cellulose biosynthesis: exciting times for a difficult field of study. Annu Rev
Plant Physiol Plant Mol Biol 50:245276
Desprez T, Juraniec M, Crowell EF, Jouy H, Pochylova Z, Parcy F, Hfte H, Gonneau M,
Vernhettes S (2007) Organization of cellulose synthase complexes involved in primary cell
wall synthesis in Arabidopsis thaliana. Proc Nat Acad Sci USA 104:1557215577
192 C. Rayon et al.
Ding S-Y, Himmel ME (2006) The maize primary cell wall microfibril: a new model derived
from direct visualization. J Agric Food Chem 54:597606
Fernandes AN, Thomas LH, Altaner CM, Callow P, Forsyth VT, Apperley DC, Kennedy CJ,
Jarvis MC (2011) Nanostructure of cellulose microfibrils in spruce wood. Proc Nat Acad Sci
USA 108:E1195E1203
Fujii S, Hayashi T, Mizuno K (2010) Sucrose synthase is an integral component of the cellulose
synthesis machinery. Plant Cell Physiol 51:294301
Fujimoto M, Arimura S, Ueda T, Takanashi H, Hayashi Y, Nakano A, Tsutsumi N (2010)
Arabidopsis dynamin-related proteins DRP2B and DRP1A participate together in clathrin-
coated vesicle formation during endocytosis. Proc Nat Acad Sci USA 107:60946099
Giddings TH Jr, Brower DL, Staehelin LA (1980) Visualization of particle complexes in the
plasma membrane of Micrasterias denticulata associated with the formation of cellulose
fibrils in primary and secondary cell walls. J Cell Biol 84:327339
Gu Y, Kaplinsky N, Bringmann M, Cobb A, Carroll A, Sampathkumar A, Baskin TI, Persson S,
Somerville C (2010) Identification of a cellulose-synthase-associated protein required for
cellulose biosynthesis. Proc Nat Acad Sci USA 107:1286612871
Gutierrez R, Lindeboom JJ, Paredez AR, Emons AMC, Ehrhardt DW (2009) Arabidopsis cortical
microtubules position cellulose synthase delivery to the plasma membrane and interact with
cellulose synthase trafficking compartments. Nat Cell Biol 11:797808
Haigler CH, Brown RM Jr (1986) Transport of rosettes from the Golgi apparatus to the plasma
membrane in isolated mesophyll cells of Zinnia elegans during differentiation to tracheary
elements in suspension culture. Protoplasma 134:111120
Herth W (1985) Plasma-membrane rosettes involved in localized wall thickening during xylem
vessel formation of Lepidium sativum L. Planta 164:1221
Himmel ME, Ding S-Y, Johnson DK, Adney WS, Nimlos MR, Brady JW, Foust TD (2007)
Biomass recalcitrance: engineering plants and enzymes for biofuels production. Science
315:804807
Hirano K, Kotake T, Kamihara K, Tsuna K, Aohara T, Kaneko Y, Takatsuji H, Tsumuraya Y,
Kawasaki S (2010) Rice BRITTLE CULM3 (BC3) encodes a classical dynamin OsDRP2B
essential for proper secondary cell wall synthesis. Planta 232:95108
Jacob-Wilk D, Kurek I, Hogan P, Delmer DP (2006) The cotton fiber zinc-binding domain of
cellulose synthase A1 from Gossypium hirsutum displays rapid turnover in vitro and in vivo.
Proc Nat Acad Sci USA 103:1219112196
Kennedy CJ, Cameron GJ, Sturcov A, Apperley DC, Altaner C, Wess TJ, Jarvis MC (2007)
Microfibril diameter in celery collenchyma cellulose: X-ray scattering and NMR evidence.
Cellulose 14:235246
Kurek I, Kawagoe Y, Jacob-Wilk D, Doblin M, Delmer D (2002) Dimerization of cotton fiber
cellulose synthase catalytic subunits occurs via oxidation of the zinc-binding domains. Proc
Nat Acad Sci USA 99:1110911114
Li S, Lei L, Somerville CR, Gu Y (2012) CSI1 links microtubules and cellulose synthase com-
plexes. Proc Nat Acad Sci USA 109:185190
Morgan JLW, Strumillo J, Zimmer J (2013) Crystallographic snapshot of cellulose synthesis and
membrane translocation. Nature 493:181186
Mueller SC, Brown RM Jr (1980) Evidence for an intramembrane component associated with a
cellulose microfibril-synthesizing complex in higher plants. J Cell Biol 84:315326
Olek AT, Rayon C, Makowski L, Kim HR, Ciesielski P, Badger J, Paul LN, Ghosh S, Kihara D,
Crowley M, Himmel ME, Bolin JT, Carpita NC (2013) Small-angle x-ray scattering reveals the
structure of the catalytic domain of a cellulose synthase and its assembly into dimers. In review
Pauly M, Keegstra K (2008) Cell-wall carbohydrates and their modification as a resource for bio-
fuels. Plant J 54:559568
Pear JR, Kawagoe Y, Schreckengost WE, Delmer DP, Stalker DM (1996) Higher plants contain
homologs of the bacterial celA genes encoding the catalytic subunit of cellulose synthase.
Proc Nat Acad Sci USA 93:1263712642
Penning BW, Hunter CT, Tayengwa R, Eveland AL, Dugard CK, Olek AT, Vermerris W, Koch
KE, McCarty DR, Davis MF, Thomas SR, McCann MC, Carpita NC (2009) Genetic
resources for maize cell wall biology. Plant Physiol 151:17031728
Potikha T, Delmer DP (1995) A mutant of Arabidopsis thaliana displaying altered patterns of
cellulose deposition. Plant J 7:453460
Roudier F, Fernandez AG, Fujita M, Himmelspach R, Borner GHH, Schindelman G, Song S,
Baskin TI, Dupree P, Wasteneys GO, Benfey PN (2005) COBRA, an Arabidopsis extracel-
lular glycosyl-phosphatidyl inositol-anchored protein, specifically controls highly aniso-
tropic expansion through its involvement in cellulose microfibril orientation. Plant Cell
17:17491763
Sato K, Suzuki R, Nishikubo N, Takenouchi S, Ito S, Nakano Y, Nakaba S, Sano Y, Funada R,
Kajita S, Kitano H, Katayama Y (2010) Isolation of a novel cell wall architecture mutant of
rice with defective Arabidopsis COBL4 ortholog BC1 required for regulated deposition of
secondary cell wall components. Planta 232:257270
Saxena IM, Brown RM Jr, Fevre M, Geremia RA, Henrissat B (1995) Multidomain architecture
of -glycosyl transferases: implications for mechanism of action. J Bacteriol 177:14191424
Schindelman G, Morikami A, Jung J, Baskin TI, Carpita NC, Derbyshire P, McCann MC, Benfey
PN (2001) COBRA encodes a putative GPI-anchored protein, which is polarly localized and
necessary for oriented cell expansion in Arabidopsis. Genes Dev 15:11151127
Sethaphong L, Haigler CH, Kubicki JD, Zimmer J, Bonetta D, Debolt S, Yingling Y (2013)
Tertiary model of a plant cellulose synthase. Proc Nat Acad Sci USA 110:75127517
Sindhu A, Langewisch T, Olek A, Multani DS, McCann MC, Vermerris W, Carpita NC, Johal G
(2007) Maize Brittle stalk2 encodes a COBRA-like protein expressed in early organ develop-
ment but required for tissue flexibility at maturity. Plant Physiol 145:14441459
Smith AM, Kruger NJ, Lunn JE (2012) Source of sugar nucleotides for starch and cellulose syn-
thesis. Proc Nat Acad Sci USA 109:E776E776
Taylor NG (2011) A role for Arabidopsis dynamin related proteins DRP2A/B in endocytosis;
DRP2 function is essential for plant growth. Plant Mol Biol 76:117129
Taylor NG, Howells RM, Huttly AK, Vickers K, Turner SR (2003) Interactions among three dis-
tinct CesA proteins essential for cellulose synthesis. Proc Nat Acad Sci USA 100:14501455
Thomas LH, Forsyth VT, Sturcov A, Kennedy CJ, May RP, Altaner CM, Apperley DC, Wess TJ,
Jarvis MC (2013) Structure of cellulose microfibrils in primary cell walls from collenchyma.
Plant Physiol 161:465476
Timmers J, Vernhettes S, Desprez T, Vincken JP, Visser RGF, Trindade LM (2009) Interactions
between membrane-bound cellulose synthases involved in the synthesis of the secondary cell
wall. FEBS Lett 583:978982
Toyooka K, Goto Y, Asatsuma S, Koizumi M, Mitsui T, Matsuoka K (2009) A mobile secre-
tory cluster involved in mass transport from the Golgi to the plant cell exterior. Plant Cell
21:12121229
Vergara CE, Carpita NC (2001) -D-Glycan synthases and the CesA gene family: lessons to
be learned from the mixed-linkage (13), (14)--D-glucan synthase. Plant Mol Biol
47:145160
Wang J, Elliott JE, Williamson RE (2008) Features of the primary wall CESA complex in wild
type and cellulose-deficient mutants of Arabidopsis thaliana. J Exp Bot 59:26272637
Zhao Q, Dixon RA (2011) Transcriptional networks for lignin biosynthesis: more complex than
we thought? Trends Plant Sci 16:227233
Zhong R, Lee C, Ye Z-H (2010) Evolutionary conservation of the transcriptional network regulat-
ing secondary cell wall biosynthesis. Trens Plant Sci 15:625632
Part III
Biomass Processing
Chapter 12
Developing Novel Enzyme Repertoires
for the Efficient Deconstruction of Plant
Biomass Tailored for the Bioenergy Industry
Harry J. Gilbert
AbstractPlant biomass, commonly referred to as lignocellulose, represents a

renewable and thus sustainable substrate for the liquid biofuel and chemical indus-
tries, which is carbon dioxide neutral. There is much debate concerning the eco-
nomic viability of lignocellulose-based liquid biofuels based, primarily, on the cost
of the enzymes required to saccharify plant biomass into its component sugars.
As a result there has been a substantial investment in enzyme technology targeted
towards improving the efficiency of plant cell wall degradation. This Chapter pro-
vides an overview of our current knowledge of plant cell wall degrading enzymes
at a structural and biochemical level. The article also describes strategies that can be
deployed to discover novel, industrially significant, enzyme functions, and how pro-
tein engineering can be used to increase the catalytic efficiency of some enzymes,
and broaden the substrate specificity of others. Finally the Chapter highlights the
emerging importance of polysaccharide oxidases in lignocellulosic deconstruction
focussing on the role these enzymes play in opening up the structure of crystalline
cellulose, explaining how they are capable of potentiating the activity of glycoside
hydrolases (cellulases) against these recalcitrant structures.
Keywords Liquid biofuels Lignocellulose glycoside hydrolases Cellulases

Mannanases Xylanases Cellulose oxidases Protein engineering
H. J. Gilbert(*)
The Institute for Cell and Molecular Biosciences, Newcastle University,
Framlington Place, Newcastle upon Tyne NE42 6EB, UK
e-mail: h.j.gilbert@ncl.ac.uk
198 H. J. Gilbert
12.1Introduction
The cell walls of plants are the most abundant source of organic carbon on the
planet. This photosynthetically fixed carbon is recycled by microbial enzymes that
convert cell wall polysaccharides to mono- and oligosaccharides. Reflecting the
growing industrial significance of plant cell wall deconstruction, the enzymes that
catalyze this process have been subjected to intense analysis in the last few years.
Today, a growing and significant application of these biocatalysts is in the produc-
tion of second generation lignocellulosic-based biofuels and in the bioprocessing
sector, where the synthesis of high-value chemicals from renewable sources, such
as plant biomass, is of particular importance (Ragauskas et al. 2006; Bayer et al.
2007; Sticklen 2008; Himmel and Bayer 2009). Plant cell walls are recalcitrant to
biological depolymerization as the extensive interactions between polysaccharides,
and between polysaccharides and lignin, restrict access to the battery of microbial
enzymes (principally glycoside hydrolases) that break down these composite struc-
tures (for review, see Mohnen 2008). Since the early 1990s, when the first structures
of these enzymes were determined, there has been an explosion of structural infor-
mation on these proteins. This structural information is starting to inform and direct
protein engineering strategies that, potentially, will make a significant contribution
to the toolbox of biocatalysts used to deconstruct the plant cell wall. These novel
enzymes could therefore increase the economic viability of using lignocellulose as
a substrate for the biofuel industry. This review will discuss how such enzymes can
be generated either through screening for natural biocatalysts that display novel, and
useful, activities, or by engineering established enzymes.
12.2The Plant Cell Wall
Plant cell walls are divided into two major types; the primary and secondary wall,
and are comprised predominantly of polysaccharides (~90%). Secondary walls,
which provide the major source of biomass, contain low amounts of pectin and
the major hemicellulose is xylan, although in gymnosperms glucomannans repre-
sent the major hemicellulose. In addition to polysaccharides, secondary walls are
often rigidified by the impregnation of lignin, a heterogenous aromatic polymer.
The structure of the plant cell has been extensively reviewed previously and will
be described briefly here (see Harris and Stone 2008; Mohnen 2008; Mohnen et al.
2008, for an overview of plant cell wall structure).
Cellulose is a -1,4-linked glucose (Glc) molecule that is substantially crystalline.
All the hemicellulosic polysaccharides contain a -linked sugar backbone. In xylans
and mannans the backbone sugars are -1,4-D-Xyl and -1,4-D-Man residues, while
in glucomannan the backbone consists of randomly dispersed -1,4-Glc and -1,4-
Man sugars. The backbones of hemicellulosic polysaccharides are decorated with a
variety of sugars and acetyl groups explaining why these polymers are not crystalline,
exemplified by xyloglucan, which comprises a -1,4-linked glucose backbone deco-
rated with xylose residues that, in turn can also contain additional sugars.
12 Developing Novel Enzyme Repertoires for the Efficient Deconstruction of Plant 199
12.3CAZy
Enzymes that modify complex carbohydrates, together with their accessory

non-catalytic carbohydrate binding modules (CBMs), have been grouped into
sequence-based families on the continuously updated CAZy database (Cantarel
et al. 2009; http://www.cazy.org/). Members of the same enzyme family dis-
play a common fold, while the catalytic apparatus and mechanism are similarly
conserved. Currently 45 of the 131glycoside hydrolase families (GHs) contain
enzymes that contribute to plant cell wall deconstruction. Of the 65 CBM families,
around half of these modules bind to components of the plant cell wall.
Cellulases: Cellulose utilization involves the activities of endo--1,4-glucanases,
cellobiohydrolases (also called exo--1,4-glucanases), and -glucosidases, that act
synergistically to convert crystalline cellulose to glucose (Beguin and Aubert 1994;
Tomme et al. 1995). Endo--1,4-glucanases attack the cellulose chain at exposed
internal glycosidic bonds, while cellobiohydrolases can attack either the non-
reducing or reducing end of exposed cellulose chains (Beguin and Aubert 1994).
Cellobiose, which can inhibit the cellobiohydrolases, is cleaved by -glucosidases
(Teeri 1997).
The substrate binding site of cellobiohydrolases consists of a tunnel through
which single cellulose chains are threaded (Rouvinen et al. 1990; Divne et al.
1994). A subset of GH9 cellulases, those containing a CBM3c, display an endo-
processive mode of action (Irwin et al. 1998; Gilad et al. 2003). CBM3cs are atyp-
ical family 3 CBMs that bind weakly to cellulose in isolation. The current model
for GH9 enzymes that contain a CBM3c is that the cellulase, through its open
substrate binding cleft, is able to bind to internal regions of a cellulose chain and
thus exhibits an endo-mode of action. However, after bond cleavage the cellulose
chain slides along the substrate binding cleft by two residues, assisted by CBM3c,
explaining its processive mode of action (Sakon et al. 1997; Irwin et al. 1998;
Gilad et al. 2003). Classically cellulose hydrolysis, deploying Hypocreajecorina
(formerly Trichoderma reesei) as the model system, is viewed as a synergistic pro-
cess between endo-acting cellulases that create new ends from that the exo-act-
ing cellobiohydrolases can release cellobiose from either the reducing (GH7 and
GH48) or non-reducing (GH6) end of the cellulose chains (reviewed in Kleywegt
et al. 1997; Teeri 1997). This model, however, is inconsistent with several features
of cellulose degradative systems. Thus biochemical and structural data indicate
that GH6 cellobiohydrolases are not, exclusively, exo acting (Amano et al. 1996;
Armand et al. 1997; Varrot et al. 1999). Furthermore, some highly active cellu-
lase systems lack a classic pair of cellobiohydrolases that act from the reducing
and non-reducing ends of cellulose chains, respectively (Xie et al. 2007; DeBoy
et al. 2008; Weiner et al. 2008). An intriguing report by Tolonen and colleagues
(2009) showed that a single endo-processive GH9 cellulase (CfCel9) was essential
for cellulose degradation in Clostridium phytofermentans. Given the redundancy
in cellulase systems, demonstration that a single enzyme is essential for cellulose
degradation is rare, and questions the classical synergy model. Indeed, CfCel9 is
an extremely exciting target in the development of consolidated bioprocessing
200 H. J. Gilbert
(CBP) systems, in which synthetic biology is deployed to introduce lignocellulosic

degradative capacity into a biofuel producing organisms, obviating the need for
external enzyme mixtures. Currently, enzyme mixtures used in the saccharifica-
tion process represents one of the most significant costs in the biofuel industry,
and thus the development of CBP organisms offers an attractive approach to gen-
erating liquid fuels. However, as detailed above, most cellulase systems consist of
a large number of enzymes, and the transfer of this degradative capacity into the
biofuel-generating organism represents a significant synthetic biology challenge.
-Mannanases: -mannanases display a (/)8 barrel-fold and are located
primarily within GH5 and GH26. Substrate recognition by -mannanases is
complicated by the requirement of these enzymes to hydrolyze the heterogenous
polymer, glucomannan. Recognition of Man and Glc sugars at subsites distal to
the active site (1 subsite) is highly variable, although some general trends are
emerging, which point to a divergence in specificity between GH5 and GH26
mannanases. GH5 mannanases are able to accommodate Glc at the 2 and +1
subsites (Tailford et al. 2009), and are thus able to hydrolyze mannosidic link-
ages flanked by Man or Glc. A particularly exciting enzyme in the context of
glucomannan degradation is Man5A from Caldanaerobius polysaccharolyticus,
which is capable of acting as both an endo-glucunase and endo-mannanase (Han
et al. 2010). With respect to glucomannan degradation Man5A is capable of ful-
filling multiple functions and thus could play an important role in the decon-
struction of softwoods.
In contrast, the GH26 mannanases generally display tight specificity for Man at
both the 2 subsite and 1 subsites. Indeed, a cohort of GH26 mannanases contain
an arginine at the 2 substrate that makes extensive interactions with the substrate,
and confers unusually high activity against small mannooligosaccharides (Ducros et
al. 2002; Cartmell et al. 2008). Screening genomic databases for other GH26 enzymes
that retain this arginine may facilitate the identification of novel mannooligosaccha-
ridases. Currently, the two Cellvibrio enzymes that contain a high-affinity 2 sub-
site do not possess additional negative binding subsites, which may explain why the
high activity displayed against mannotriose and mannotetraose is not translated to
the hydrolysis of polysaccharides (Hogg et al. 2001; Cartmell et al. 2008). However,
these enzymes provide an excellent structural scaffold for building additional distal
subsites that, when coupled with the very high affinity 2 subsite, has the potential to
generate mannanases with extremely high catalytic activities.
12.4Structural Changes that Modulate the Mode of

Enzyme Action
The structural basis for the GH6 and GH7 cellobiohydrolases and endoglucanases,
reflecting the formation of extended loops that form tunnel-like substrate bind-
ing regions, is well established and has been extensively reviewed (Kleywegt et
al. 1997; Teeri 1997; Varrot et al. 1999). However, it is also apparent that subtle
changes to the distal negative subsites of glycoside hydrolases can be used modu-
late the mode of enzyme action. Thus, endo-acting GH43 arabinanases contain a
substrate binding cleft open at both ends, explaining their endo-activity (Alhassid
et al. 2009), The single arabinanase from Cellvibrio japonicus, CjArb43A, unu-
sually, displays an endo-processive activity (McKie et al. 1997). Again sub-
tle changes to the distal subsite (compared to endo-acting arabinanases) cause a
steric block that prevents extension of the substrate past the 3 subsite (Nurizzo
et al. 2002; Proctor et al. 2005). Endo-activity can be introduced into the enzyme
through only two amino acid substitutions to the 3 subsite, without compromis-
ing catalytic efficiency. The redesigned endo-enzyme remains a more powerful
catalyst than unmodified endo-acting arabinanases, demonstrating the feasibility
of engineering industrially relevant modes of action into highly active carbohy-
drate modifying enzymes.
12.5Xylan Degradation
The xylan backbone is hydrolysed primarily by GH10 and GH11 xylanases, while
the Araf side chains are removed by enzymes from various families including
GH43, GH51, GH54 and GH62 (for review of xylan degradation, see Gilbert et
al. 2008). GH43 arabinofuranosidases may display the highest level of substrate
diversity exemplified by the activity of a GH43 enzyme (HiAXHd3) that removes
the O3- Araf side chain from Xyl residues decorated at both O2 and O3 with Araf
(van den Broek et al. 2005; Sorensen et al. 2006). The crystal structure of this
enzyme reveals a cleft that houses the xylan backbone. In the centre of the cleft is
the active site pocket. Modification of the rim of the active site pocket introduces
endo-xylanase activity, while the resultant enzyme variant, Y165A, retains arab-
inofuranosidase activity. The crystal structure of Y165A shows that the mutation
creates a topology that allows either the xylan backbone or arabinose side chains
to enter the active site, explaining the observed multiple catalytic functions of the
engineered enzyme, Fig.12.1 (McKee et al. 2012). These data demonstrate that
the active site of HiAXHd3 is tuned to hydrolyse arabinofuranosyl or xylosyl link-
ages, and it is the topology of the distal regions of the substrate binding surface
that confers specificity. The introduction of xylanase activity into an AXHd3 ara-
binofuranosidase is of considerable biotechnological significance, particularly in
industries, such as the biofuel sector, which utilize the plant cell wall as the major
substrate. The chemistry of plant cell walls is highly complex, and thus a large
number of enzymes, with different substrate specificities, are required to fully sac-
charify these composite structures. From an industrial perspective, the enzyme
component in the bioenergy and bioprocessing sectors is a significant cost, and
thus it is important to minimize the number of different glycoside hydrolases
deployed. By introducing additional catalytic functions into a biotechnologically
significant glycoside hydrolase, the work of McKee et al. demonstrates the fea-
sibility of generating limited enzyme cocktails that display the range of activities
202 H. J. Gilbert
Fig.12.1An engineered arabinofuranosidase that is able to degrade xylan. Panel a Orthogonal

views of the active site cleft of the wild type HiAXHd3 arabinofuranosidase shown as a molecular
surface with the sugar ligand, coloured green (Xyl) or yellow (Ara) shown as sticks. The position of
Y165 is shown in blue. Panel b The active site cleft of The Y166A mutant shown as for panel a, the
removal of the Y166 side-chain increases the accessibility of the active site to xylan. Panel c shows
the catalytic activity of wild type and the Y165A/F493A/E593A mutant of HiAXHd3
required to efficiently degrade plant cell walls. In a more generic sense the engi-
neering of HiAXHd3 demonstrates that GH43 glycoside hydrolases provide a
platform for generating bespoke multi-functional enzymes that target industrially
significant, chemically complex, substrates, exemplified by the plant cell wall.
12.6How Do We Look for Novel Plant Cell Wall Degrading

Enzymes?
One of the major challenges facing the liquid biofuels industry is the cost of
enzymes used in the process. As a result there has been a significant investment in
enzyme discovery, in the hope of finding novel and superior (to current enzymes)
plant cell wall degrading biocatalysts. Two potential strategies for mining exist-
ing genomic data is to identify proteins with a non-catalytic module that indicates
a plant cell wall degrading function, but, also contains a domain that is >150
amino acids, which may comprise a novel enzyme. Recently, Bras et al. (2011)
characterized a protein that is expressed at highly levels when Clostridium ther-
emocellum is cultured on cellulose. The protein contains a ~250 residue sequence
with no sequence similarity with any entries in the CAZy database, and a type I
dockerin, a domain that integrates proteins into the cellulosome, a highly efficient
plant cell wall degrading multienzyme complex. Analysis of the protein showed it
to be an endo--1,4-glucanase that acts in synergy with the cellulosomal cellobio-
hydrolase. The crystals structure of the enzyme (Cel124A) in complex with cel-
lotriose showed that the glycanase hydrolyzed glycosidic bonds at the interface
between crystalline and amorphous regions of cellulose. Indeed, the specificity of

Cel124 for precise structures within cellulose may explain why C. thermocellulm
expresses such a large number of different endoglucanases. It is possible that the
enzymes recognize different substructures of cellulose, which may explain why
the cellulosome is one of the most efficient crystalline cellulose degrading sys-
tems known.
In addition to searching for non-CAZy sequences, enzymes with novel spe-
cificities can also be identified by interrogating the CAZy database for proteins
that display features that are atypical of the family in which they are located. This
approach has led the Fontes laboratory to identify a GH5 enzyme in which there is
departure from the highly conserved amino acids that interact with O3 of the sugar
bound in the active site. The enzyme was characterized and shown to be an arabi-
noxylan specific xylanase, in which the specificity was conferred by a pocket that
was linked -1,3 to the xylose in the active site (Correia et al. 2011, Montanier
et al. 2011). Thus, the productive binding energy to the arabinose, compensated
for the loss of interactions with O3 of the xylose bound in the active site, explain-
ing why the enzyme only hydrolyzes arabinoxylan and not undecorated xylan.
Significantly, the distal region of the substrate binding cleft is unusually open sug-
gesting that the enzyme is capable of attacking highly decorated xylans. This has
now been confirmed; the enzyme is able to attack corn stem xylan, an extremely
complex form of the hemicelluloses, while typical GH10 and GH11 xylanases dis-
play no activity against this polysaccharide.
12.7Plant Cell Wall Degrading Enzymes Display Complex

Molecular Architectures
Microbial plant cell wall hydrolases display complex molecular architectures in

which the catalytic module is appended, by flexible linker sequences, to one or
more CBMs (reviewed in Boraston et al. 2004). CBMs have now been described
that bind to the major polysaccharides found in plant cell wall structures (for
review see Boraston et al. 2004). Although ubiquitous, the mechanism by which
CBMs potentiate catalysis remains unclear, the most likely explanation is that
they reduce the accessibility problem by simply bringing the appended catalytic
modules into intimate association with their target substrate. Proteins that dis-
play a CBM-like function, termed Expansins, may offer advantages over classic
CBMs. These proteins have been shown to mechanically weaken plant cell walls
(McQueen-Mason and Cosgrove 1994), and their use in improving cellulase effi-
ciency has been reported (Han and Chen 2007). Currently, expansins appear to
disrupt the cellulose-hemicellulose interface, while the functional importance of
Swollenin, another cellulose binding protein, remains opaque.
Type B CBMs generally bind to substrates of the catalytic modules. Exceptions
to this rule include the CBM35s appended to three xylan degrading enzymes,
which bind to both glucuronic acid (GlcA) and the unsaturated product released
204 H. J. Gilbert
Fig.12.2Model for the role of BsCBM66 in exo-acting -fructosidases. The binding of

BsCBM66 to terminal fructofuranose residues enhances the activity of two GH32 broad-acting
-fructosidases against levan, but not against the other major fructan inulin. It is proposed that
BsCBM66 facilitates the enzymatic targeting of branched substrates, such as levan, by binding
in synergy to the terminal fructose residues of a branch structure leading to increased affinity
through avidity effects. This avidity effect does not occur when the enzyme is attacking linear
fructans such as inulin
by pectate lyases, but not to 4-O-methyl-D-glucuronic acid (MeGlcA), the more

common uronic acid found in xylans. It has been proposed that the rate of GlcA
methylation is lower than glucuronoxylan synthesis in rapidly dividing cells (Pea
et al. 2007). This has led to the hypothesis that, by targeting unmethylated uronic,
the CBM is directing enzymes to more open structures that are particularly sus-
ceptible to enzyme degradation. It is possible that this cohort of CBM35s initially
direct the xylan degrading apparatus to regions of cell walls that are being actively
degraded, for which anhydrogalacturonic is a marker, but, as xylan structures are
revealed, the enzyme is shuttled onto the hemicellulosic polysaccharide affording
the enzyme access to its target substrate (Montanier et al. 2009), a view consistent
with recent data showing that these CBMs enhance degradation of non-methylated
glucuronoxylans (Urbanowicz et al. 2012).
Recent studies by Cuskin et al. (2012) have shown that members of a new
CBM family, CBM66, are exo binders recognizing the non-reducing end of
branched fructan polysaccharides. When an exo-acting glycoside hydrolase that
attacks fructans was linked to the CBM66 there was a 100-fold enhancement in
activity again levan, a highly branched fructans, but not against inulin, a linear
undecorated fructan. These data show that the CBM66 enhances catalytic activity
by increasing the affinity of the enzyme for its substrate through an avidity effect
(Fig.12.2). Such a mechanism can only occur if the CBM and the appended cat-
alytic module are binding to the same polysaccharide molecule, explaining why
the potentiation only occurs for branched substrates. Within CBM66 are mem-
bers appended to arabinofuranosidase families (GH43 and GH51), and thus this
CBM-mediated targeting of branched polysaccharides is highly relevant to plant

cell wall degradation. Indeed, the technology developed around the CBM66 fam-
ily provides a platform for engineering enzymes exposed to different evolutionary
pressures to CBM-containing glycoside hydrolases.
12.8Overcoming the Problem of Crystalline Cellulose with

Cellulose Oxidases
The major barrier to deconstructing cellulose, the most abundant substrate available
to the biofuel industry, is its highly crystalline structure. Glycoside hydrolases dis-
tort the sugar in the active site into its transition state conformation, and to achieve
this substrate distortion, enzymes need to bind isolated chains of their target poly-
saccharides. As cellulose chains form highly crystalline structures it is not obvious
how cellulases are able to channel isolated cellulose chains into their active sites.
It has been shown by Koivula et al. (1998) that a tryptophan, at the entrance to the
active site of the pivotal Trichoderma cellobiohydrolase, Cel6A, is essential for
activity against crystalline cellulose, but is not required when the enzyme is acting
on disordered or soluble substrates. It was proposed that the tryptophan intercalates
between a cellulose microfibril and a surface cellulose chain, and the resultant sol-
ubilised glucan molecule can then be fed into the active site tunnel.
A more generic mechanism for attacking crystalline polysaccharides is the
recent discovery of polysaccharide oxidases. These oxidases were first shown to
oxidize and thus disrupt the crystalline structure of chitin, making the polysaccha-
ride highly accessible to chitinases (glycoside hydrolases), and thus greatly poten-
tiating the activity of the glycoside hydrolases (Vaaje-Kolstad et al. 2005, 2010).
The role of oxidases has now been extended to cellulose where cellulose oxidases
mediate a similar disruptive oxidation of the polysaccharide, increasing its access
to cellulase action (Forsberg et al. 2011; Quinlan et al. 2011). These enzymes con-
tain a copper binding site, and the redox metal plays a central role in the oxidation
reaction (Quinlan et al. 2011; Aachmann et al. 2012; Vaaje-Kolstad et al. 2012).
Significantly, the oxidation reaction does not require substrate distortion and thus
the enzymes can bind to the planar surface presented by crystalline polysaccha-
rides such as cellulose, explaining their unique activity against high recalcitrant
substrates. These oxidases represent a major advance in cellulose deconstruction,
and comprises the disruptive C1 factor proposed by Reese and colleagues (1950).
It should be noted that while these oxidases play a critical role in the cellulases
systems of both aerobic fungi (Harris et al. 2010; Quinlan et al. 2012) and bacte-
ria (Forsberg et al. 2011), they are absent in the corresponding anaerobic cellulase
complexes. Despite the lack of oxidative enzymes anaerobic cellulase consortia,
exemplified by the Clostridium thermocellum cellulosomes (Fontes and Gilbert
2010), display similar catalytic efficiencies against crystalline substrates to the
corresponding aerobic fungal systems (Ding et al. 2012). How the anaerobic cel-
lulase systems overcome the lack of an oxidative enzyme is currently unclear.
206 H. J. Gilbert
12.9Future Perspectives
It is evident that the explosion of omics technologies, genomic, metagenomic and

metatranscriptomic, provides us with an unrivalled opportunity for enzyme discov-
ery programmes. Indeed the power of this information is revealed by the important
recent discovery of polysaccharide oxidases. In the next few years the smart analy-
sis of omics data, in conjunction with the further development of oxidases, and the
continued use of structure based rational design, is likely to generate highly effi-
cient lignocellulosic degrading enzyme cocktails. This portfolio of biocatalysts is
likely to remove the rate limiting step in the use of plant biomass as an economi-
cally viable substrate for biofuel production.
References
Aachmann FL, Sorlie M, Skjak-Brak G, Eijsink VGH, Vaaje-Kolstad G (2012) NMR structure of
a lytic polysaccharide monooxygenase provides insight into copper binding, protein dynam-
ics, and substrate interactions. Proc Natl Acad Sci USA 109:1877918784
Alhassid A, Ben-David A, Tabachnikov O, Libster D, Naveht E, Zolotnitsky G, Shoham Y,
Shoham G (2009) Crystal structure of an inverting GH 43 1,5-alpha-L-arabinanase from
Geobacillus stearothermophilus complexed with its substrate. Biochem J 422:7382
Amano Y, Shiroishi M, Nisizawa K, Hoshino E, Kanda T (1996) Fine substrate specificities of
four exo-type cellulases produced by Aspergillus niger, Trichoderma reesei, and Irpex lac-
teus on (13), (14)--D-glucans and xyloglucan. J Biochem 120:11231129
Armand S, Drouillard S, Schulein M, Henrissat B, Driguez H (1997) A bifunctionalized fluoro-
genic tetrasaccharide as a substrate to study cellulases. J Biol Chem 272:27092713
Bayer EA, Lamed R, Himmel ME (2007) The potential of cellulases and cellulosomes for cellu-
losic waste management. Curr Opin Biotechnol 18:237245
Beguin P, Aubert JP (1994) The biological degradation of cellulose. FEMS Microbiol Rev
13:2558
Boraston AB, Bolam DN, Gilbert HJ, Davies GJ (2004) Carbohydrate-binding modules: fine-tun-
ing polysaccharide recognition. Biochem J 382:769781
Bras JL, Cartmell A, Carvalho AL, Verze G, Bayer EA, Vazana Y, Correia MA, Prates JA,
Ratnaparkhe S, Boraston AB, Romao MJ, Fontes CM, Gilbert HJ (2011) Structural insights
into a unique cellulase fold and mechanism of cellulose hydrolysis. Proc Natl Acad Sci USA
108:52375242
Cantarel BL, Coutinho PM, Rancurel C, Bernard T, Lombard V, Henrissat B (2009) The
Carbohydrate-Active EnZymes database (CAZy): an expert resource for glycogenomics.
Nucl Acids Res 37:233238
Cartmell A, Topakas E, Ducros VM, Suits MD, Davies GJ, Gilbert HJ (2008) The Cellvibrio
japonicus mannanase CjMan26C displays a unique exo-mode of action that is conferred by
subtle changes to the distal region of the active site. J Biol Chem 283:3440334413
Correia MA, Mazumder K, Bras JL, Firbank SJ, Zhu Y, Lewis RJ, York WS, Fontes CM,
Gilbert HJ (2011) Structure and function of an arabinoxylan-specific xylanase. J Biol Chem
286:2251022520
Cuskin F, Flint JE, Gloster TM, Morland C, Basle A, Henrissat B, Coutinho PM, Strazzulli A,
Solovyova AS, Davies GJ, Gilbert HJ (2012) How nature can exploit nonspecific catalytic
and carbohydrate binding modules to create enzymatic specificity. Proc Natl Acad Sci USA
109:2088920894
DeBoy RT, Mongodin EF, Fouts DE, Tailford LE, Khouri H, Emerson JB, Mohamoud Y, Watkins
K, Henrissat B, Gilbert HJ, Nelson KE (2008) Insights into plant cell wall degradation from
the genome sequence of the soil bacterium Cellvibrio japonicus. J Bacteriol 190:54555463
Ding SY, Liu YS, Zeng Y, Himmel ME, Baker JO, Bayer EA (2012) How does plant cell wall
nanoscale architecture correlate with enzymatic digestibility? Science 338:10551060
Divne C, Stahlberg J, Reinikainen T, Ruohonen L, Pettersson G, Knowles JK, Teeri TT, Jones
TA (1994) The three-dimensional crystal structure of the catalytic core of cellobiohydrolase I
from Trichoderma reesei. Science 265:524528
Ducros VM, Zechel DL, Murshudov GN, Gilbert HJ, Szabo L, Stoll D, Withers SG, Davies GJ
(2002) Substrate distortion by a -mannanase: snapshots of the Michaelis and covalent-inter-
mediate complexes suggest a B(2,5) conformation for the transition state. Angew Chem Int
Ed Engl 41:28242827
Fontes CM, Gilbert HJ (2010) Cellulosomes: highly efficient nanomachines designed to decon-
struct plant cell wall complex carbohydrates. Annu Rev Biochem 79:655681
Forsberg Z, Vaaje-Kolstad G, Westereng B, Bunaes AC, Stenstrom Y, MacKenzie A, Sorlie
M, Horn SJ, Eijsink VGH (2011) Cleavage of cellulose by a CBM33 protein. Protein Sci
20:14791483
Gilad R, Rabinovich L, Yaron S, Bayer EA, Lamed R, Gilbert HJ, Shoham Y (2003) CelI, a
noncellulosomal family 9 enzyme from Clostridium thermocellum, is a processive endoglu-
canase that degrades crystalline cellulose. J Bacteriol 185:391398
Gilbert HJ, Stalbrand H, Brumer H (2008) How the walls come crumbling down: recent struc-
tural biochemistry of plant polysaccharide degradation. Curr Opin Plant Biol 11:338348
Han YJ, Chen HZ (2007) Synergism between corn stover protein and cellulase. Enzyme Microb
Technol 41:638645
Han Y, Dodd D, Hespen CW, Ohene-Adjei S, Schroeder CM, Mackie RI, Cann IK (2010)
Comparative analyses of two thermophilic enzymes exhibiting both -1,4 mannosidic and
-1,4 glucosidic cleavage activities from Caldanaerobius polysaccharolyticus. J Bacteriol
192:41114121
Harris PJ, Stone BA (2008) Chemistry and molecular organization of plant cell walls. In:
Himmel ME (ed) Biomass recalcitrance: deconstructing the plant cell wall for bioenergy.
Blackwell, Oxford, pp 6093
Harris PV, Welner D, McFarland KC, Re E, Poulsen J-CN, Brown K, Salbo R, Ding H, Vlasenko
E, Merino S, Xu F, Cherry J, Larsen S, Lo Leggio L (2010) Stimulation of lignocellulosic
biomass hydrolysis by proteins of glycoside hydrolase family 61: structure and function of a
large, enigmatic family. Biochemistry 49:33053316
Himmel ME, Bayer EA (2009) Lignocellulose conversion to biofuels: current challenges, global
perspectives. Curr Opin Biotechnol 20:316317
Hogg D, Woo EJ, Bolam DN, McKie VA, Gilbert HJ, Pickersgill RW (2001) Crystal structure of
mannanase 26A from Pseudomonas cellulosa and analysis of residues involved in substrate
binding. J Biol Chem 276:3118631192
Irwin D, Shin DH, Zhang S, Barr BK, Sakon J, Karplus PA, Wilson DB (1998) Roles of the cata-
lytic domain and two cellulose binding domains of Thermomonospora fusca E4 in cellulose
hydrolysis. J Bacteriol 180:17091714
Kleywegt GJ, Zou JY, Divne C, Davies GJ, Sinning I, Stahlberg J, Reinikainen T, Srisodsuk
M, Teeri TT, Jones TA (1997) The crystal structure of the catalytic core domain of endo-
glucanase I from Trichoderma reesei at 3.6 resolution, and a comparison with related
enzymes. J Mol Biol 272:383397
Koivula A, Kinnari T, Harjunpaa V, Ruohonen L, Teleman A, Drakenberg T, Rouvinen J, Jones
TA, Teeri TT (1998) Tryptophan 272: an essential determinant of crystalline cellulose degra-
dation by Trichoderma reesei cellobiohydrolase Cel6A. FEBS Lett 429:341346
McKee LS, Pena MJ, Rogowski A, Jackson A, Lewis RJ, York WS, Krogh KB, Vikso-Nielsen A,
Skjot M, Gilbert HJ, Marles-Wright J (2012) Introducing endo-xylanase activity into an exo-
acting arabinofuranosidase that targets side chains. Proc Natl Acad Sci USA 109:65376542
208 H. J. Gilbert
McKie VA, Black GW, Millward-Sadler SJ, Hazlewood GP, Laurie JI, Gilbert HJ (1997)
Arabinanase A from Pseudomonas fluorescens subsp. cellulosa exhibits both an endo- and an
exo- mode of action. Biochem J 323:547555
McQueen-Mason S, Cosgrove DJ (1994) Disruption of hydrogen bonding between plant
cell wall polymers by proteins that induce wall extension. Proc Natl Acad Sci USA
91:65746578
Mohnen D (2008) Pectin structure and biosynthesis. Curr Opin Plant Biol 11:266277
Mohnen D, Bar-Peled M, Somerville C (2008) Biosynthesis of plant cell walls. In: Himmel
ME (ed) Biomass recalcitrance: deconstructing the plant cell wall for bioenergy. Blackwell,
Oxford, pp 94187
Montanier C, van Bueren AL, Dumon C, Flint JE, Correia AM, Prates JA, Firbank SJ, Lewis
RJ, Grondin GG, et al (2009) Evidence that family 35 carbohydrate binding modules display
conserved specificity but divergent function. Proc Natl Acad Sci USA 106:30653070
Montanier CY, Correia MA, Flint JE, Zhu Y, Basle A, McKee LS, Prates JA, Polizzi SJ, Coutinho
PM, Lewis RJ, Henrissat B, Fontes CM, Gilbert HJ (2011) A novel, noncatalytic carbohy-
drate-binding module displays specificity for galactose-containing polysaccharides through
calcium-mediated oligomerization. J Biol Chem 286:2249922509
Nurizzo D, Turkenburg JP, Charnock SJ, Roberts SM, Dodson EJ, McKie VA, Taylor EJ, Gilbert
HJ, Davies GJ (2002) Cellvibrio japonicus -L-arabinanase 43A has a novel five-blade
-propeller fold. Nat Struct Biol 9:665668
Pea MJ, Zhong R, Zhou GK, Richardson EA, ONeill MA, Darvill AG, York WS, Ye ZH (2007)
Arabidopsis irregular xylem8 and irregular xylem9: implications for the complexity of glucu-
ronoxylan biosynthesis. Plant Cell 19:549563
Proctor MR, Taylor EJ, Nurizzo D, Turkenburg JP, Lloyd RM, Vardakou M, Davies GJ, Gilbert
HJ (2005) Tailored catalysts for plant cell-wall degradation: redesigning the exo/endo prefer-
ence of Cellvibrio japonicus arabinanase 43A. Proc Natl Acad Sci USA 102:26972702
Quinlan RJ, Sweeney MD, Lo Leggio L, Otten H, Poulsen JC, Johansen KS, Krogh KBRM,
Jorgensen CI, Tovborg M, Anthonsen A, Tryfona T, Walter CP, Dupree P, Xu F, Davies GJ,
Walton PH (2011) Insights into the oxidative degradation of cellulose by a copper metalloen-
zyme that exploits biomass components. Proc Natl Acad Sci USA 108:1507915084
Ragauskas AJ, Williams CK, Davison BH, Britovsek G, Cairney J, Eckert CA, Frederick WJ Jr,
Hallett JP, Leak DJ, Liotta CL, Mielenz JR, Murphy R, Templer R, Tschaplinski T (2006)
The path forward for biofuels and biomaterials. Science 311:484489
Reese ET, Siu RG, Levinson HS (1950) The biological degradation of soluble cellulose deriva-
tives and its relationship to the mechanism of cellulose hydrolysis. J Bacteriol 59:485497
Rouvinen J, Bergfors T, Teeri T, Knowles JK, Jones TA (1990) Three-dimensional structure of
cellobiohydrolase II from Trichoderma reesei. Science 249:380386
Sakon J, Irwin D, Wilson DB, Karplus PA (1997) Structure and mechanism of endo/exocellulase
E4 from Thermomonospora fusca. Nat Struct Biol 4:810818
Sorensen HR, Jorgensen CT, Hansen CH, Jorgensen CI, Pedersen S, Meyer AS (2006) A novel
GH43 -L-arabinofuranosidase from Humicola insolens: mode of action and synergy with
GH51 -L-arabinofuranosidases on wheat arabinoxylan. Appl Microbiol Biotechnol 73:850861
Sticklen MB (2008) Plant genetic engineering for biofuel production: towards affordable cellu-
losic ethanol. Nat Rev Genet 9:433443
Tailford LE, Ducros VM, Flint JE, Roberts SM, Morland C, Zechel DL, Smith N, Bjornvad
ME, Borchert TV, Wilson KS, Davies GJ, Gilbert HJ (2009) Understanding how diverse
-mannanases recognize heterogeneous substrates. Biochemistry 48:70097018
Tomme P, Warren RA, Gilkes NR (1995) Cellulose hydrolysis by bacteria and fungi. Adv Microb
Physiol 37:181
Teeri TT (1997) Crystalline cellulose degradation: New insight into the function of cellobiohy-
drolases. Trends Biotechnol 15:160167
Tolonen AC, Chilaka AC, Church GM (2009) Targeted gene inactivation in Clostridium phyto-
fermentans shows that cellulose degradation requires the family 9 hydrolase Cphy3367. Mol
Microbiol 74:13001313
Urbanowicz BR, Pena MJ, Ratnaparkhe S, Avci U, Backe J, Steet HF, Foston M, Li H, ONeill
MA, Ragauskas AJ, Darvill AG, Wyman C, Gilbert HJ, York WS (2012) 4-O-methylation of
glucuronic acid in Arabidopsis glucuronoxylan is catalyzed by a domain of unknown func-
tion family 579 protein. Proc Natl Acad Sci USA 109:1425314258
Vaaje-Kolstad G, Horn SJ, van Aalten DMF, Synstad B, Eijsink VGH (2005) The non-catalytic
chitin-binding protein CBP21 from Serratia marcescens is essential for chitin degradation. J
Biol Chem 280:2849228497
Vaaje-Kolstad G, Westereng B, Horn SJ, Liu Z, Zhai H, Sorlie M, Eijsink VGH (2010) An oxi-
dative enzyme boosting the enzymatic conversion of recalcitrant polysaccharides. Science
330:219222
Vaaje-Kolstad G, Bohle LA, Gaseidnes S, Dalhus B, Bjoras M, Mathiesen G, Eijsink VGH
(2012) Characterization of the chitinolytic machinery of Enterococcus faecalis V583 and
high-resolution structure of its oxidative CBM33 enzyme. J Mol Biol 416:239254
van den Broek LA, Lloyd RM, Beldman G, Verdoes JC, McCleary BV, Voragen AG (2005)
Cloning and characterization of arabinoxylan arabinofuranohydrolase-D3 (AXHd3) from
Bifidobacterium adolescentis DSM20083. Appl Microbiol Biotechnol 67:641647
Varrot A, Schulein M, Davies GJ (1999) Structural changes of the active site tunnel of
Humicola insolens cellobiohydrolase, Cel6A, upon oligosaccharide binding. Biochemistry
38:88848891
Weiner RM, Taylor LE 2nd, Henrissat B, Hauser L, Land M, Coutinho PM, Rancurel C,
Saunders EH, Longmire AG, Zhang H, Bayer EA, Gilbert HJ, Larimer F, Zhulin IB, Ekborg
NA, Lamed R, Richardson PM, Borovok I, Hutcheson S (2008) Complete genome sequence
of the complex carbohydrate-degrading marine bacterium, Saccharophagus degradans strain
2-40 T. PLoS Genet 4:e1000087
Xie G, Bruce DC, Challacombe JF, Chertkov O, Detter JC, Gilna P, Han CS, Lucas S, Misra M,
Myers GL, Richardson P, Tapia R, Thayer N, Thompson LS, Brettin TS, Henrissat B, Wilson
DB, McBride MJ (2007) Genome sequence of the cellulolytic gliding bacterium Cytophaga
hutchinsonii. Appl Environ Microbiol 73:35363546
Chapter 13
Using Natural Plant Cell Wall Degradation
Mechanisms to Improve Second Generation
Bioethanol
Adriana Grandis, Amanda P. de Souza, Eveline Q. P. Tavares

and Marcos S. Buckeridge
AbstractCell wall hydrolysis is one of the key processes needed for develop-
ment of the technology for second-generation (2G) bioethanol production. Thus,
finding and characterizing enzymes that can deal with the complexity of the walls
has been the main focus of research. As a result, data on pretreatments of many
kinds and performances of enzyme cocktails containing glycosyl hydrolases from
microorganisms are becoming quickly available. Here we propose that the effi-
ciency of the 2G process could be increased even further by acquiring control of
mechanisms that plants themselves use to degrade their own walls, so that wall
loosening provoked by such processes would decrease the energy demand for pre-
treatments and facilitate hydrolysis. The examined in this chapter are plant-micro-
organism interaction, cell wall storage mobilization, fruit ripening, abscission, and
aerenchyma formation. These systems are seen as having in common the use of
modules that are coupled sequentially in order to lead to cell wall modification,
including hydrolysis, for performance of different biological functions. These
modules are (1) target cells perception of a message from the hormonal balance,
(2) cell separation, (3) cell expansion, (4) programmed cell death, (5) hemicellu-
lose-cellulose relaxation/hydrolysis and (6) cellulose hydrolysis. We propose that
the use of synthetic biology to transform bioenergy feedstocks could be a route to
increase the efficiency of 2G processes.
13.1Introduction
The main bioethanol producers in the world are the US and Brazil, the former pro-
ducing it from maize starch and the second from sucrose from sugarcane culms.
In order to increase bioethanol production without increasing acreage, second
A. Grandis A. P. de Souza E. Q. P. Tavares M. S. Buckeridge(*)

Laboratory of Plant Physiological Ecology, Department of Botany, Institute of Biosciences,
University of So Paulo, So Paulo, Brazil
e-mail: msbuck@usp.br
212 A. Grandis et al.
generation (2G) bioethanol technologies could be used. These technologies con-

sist in the production of free fermentable sugars from cell walls. To access these
sugars it is of great importance the production of basic scientific information about
what the cell walls are like and how enzymes attack each polymer as well as the
entire composite.
The production of bioethanol by means of 2G technologies involves the use of
strategies that should modify the organization of cell wall architecture. The so-
called 2G route involves biomass pretreatments in which polysaccharides would
become available for enzymatic hydrolysis by application of physical and/or
chemical treatments (Soccol et al. 2010; Dos Santos et al. 2011).
Moreover, due to the complex nature of the plant cell wall, degradation of lig-
nocellulosic biomass requires the use of different classes of enzymes that have to
be used in high proportions in order to produce enough free sugars for efficient
fermentation (Verma et al. 2010). The production of 2G bioethanol requires ca.
11 millions of filter paper units (FPU) (i.e. 19kg) to produce 84L of bioethanol
(Himmel et al. 1997, 1999) or 1525kg of cellulase per ton of biomass (Carroll
and Somerville 2009; Taylor et al. 2008). Furthermore, due to the different compo-
sitions of the cell wall polymers, it would still be necessary to prospect for individ-
ually different classes of enzymes for characterization and subsequently combine
them into cocktails that should be specific for every type of biomass, including
the pretreated ones. Thus, one of the first challenges to be faced in order to turn
2G bioethanol viable commercially is to develop efficient systems of enzymes for
biomass degradation (Verma et al. 2010). Due to the high costs and limited capac-
ity for enzyme production, new ways to make biomass available for fermentation
could be helpful, since they could hopefully lead to the use of lower proportions of
enzymes in the process.
The proposed way to couple pretreatment and hydrolysis has been the use of
enzyme cocktails on cell walls of pretreated materials (Fig.13.1). Such enzyme
cocktails are usually artificial assemblies of extracellular proteins produced by
microorganisms (Balat 2011). In fact, many initiatives exist to use microorgan-
isms as producers of heterologous glycosyl hydrolases so that enzyme engineering
could be used to improve their action.
The sources of enzymes chosen by the scientific community varies, but with
the advent of metagenomic techniques, microorganisms and animal genes that
encode for glycosyl hydrolases became the main focus for enzyme search in the
hope to find new targets that could be somehow more efficient to hydrolyze plant
cell walls. This strategy relies mainly on the idea that microorganisms that per-
form composting of plant residues in the environment and some animal digestive
systems are capable to hydrolyze the plant cell walls (Gmez et al. 2008).
Some initiatives exist in which enzymes coding genes from microorganisms
have been cloned into plants in order to try to loosen walls and turn them more
amenable to hydrolysis (Kaida et al. 2009; Xin et al. 2011). Although expres-
sion of the enzymes in vivo have been demonstrated, there is no evidence that
these biomasses would be more easily hydrolysed and/or pretreated than biomass
derived from untransformed plants.
13 Using Natural Plant Cell Wall Degradation Mechanisms 213
Fig.13.1Schematic representation of the second-generation process of bioethanol production,

highlighting the different types of pretreatments and hydrolysis
Thus, even with the large effort of the scientific community world wide to
improve pretreatment and hydrolysis processes, cell wall recalcitrance to hydroly-
sis continues to be a barrier to turn the 2G processes economically viable (Himmel
et al. 2007). A common mistake has been to attribute recalcitrance solely to the
presence of lignin. However, this recalcitrance is also due to the enormous com-
plexity of the architecture of the cell wall with its several polymers integrated in a
composite that has been selected in a way to avoid enzyme attack and degradation.
One additional strategy to be used to improve the knowledge about 2G pro-
cesses that has not been consistently thought of, or experimented with the same
intensity, is the use of hydrolytic systems that plants themselves possess. Some of
them can be quite efficient to degrade and/or change the architecture of cell walls
and might lead to important additions to 2G processes, especially with the advent
of the synthetic biology era, in which the capacity of changing biological systems
(e.g. turning on/off entire biochemical pathways) is expect to be developed.
Similarly to the strategies currently being designed in which biological engi-
neering is going to be used within microorganisms to produce more efficient
enzymes, plant biotechnology could help not only by the introduction of genes
of animals and/or microorganisms, but also by finding mechanisms of cell wall
hydrolysis that already exist in nature. The main mechanisms of this kind known
are: (1) plant-microorganism interaction; (2) seed cell wall storage mobilization;
(3) fruit ripening, (4) abscission and (5) aerenchyma formation.
In this chapter, we will provide a review of aspects of some representative
mechanisms that could be relevant to improve plants so that they would be better
prepared for pretreatment and hydrolysis. Here, this process will be named bio-
logical pretreatment. The idea is that in the future, plants could have some of
their hydrolytic processes redesigned to alter some features of the cell walls in the
way of turning them more suitable for use in industry, without however loosing
the perspective of the technologies current under development for 2G bioethanol
production.
In order to understand how walls can be hydrolysed, some aspects of cell wall
composition, structure and architecture will be introduced in the following section.
The understanding of the cell wall complexity will allow the reader to appreciate
how some natural systems in which walls are modified could be used for biotech-
nology applications in bioethanol production.
13.2Composition, Structure and Architecture of the Plant

Cell Wall
The cell walls are composed of a mixture of polymers, mainly carbohydrates, pro-
teins and secondary metabolites. These polymers interact through a mixture of
covalent and non-covalent linkages to form a supramolecular complex that is
thought to be a multifunctional structure responsible for controlling the mechani-
cal properties of the cell.
In order to elucidate plant cell wall structure and functions, wall scientists have
followed some key steps mainly during the second half of the 20th Century. The
first step was to set up methods for extraction, determination of composition and
structure of wall components. This step was followed by the development of meth-
ods to determine the structure (linkage types) and physico-chemical properties
of these polymers. All this work culminated in the proposition of a succession of
models that pictured a general assembly for the polymers in the wall (Keegstra
et al. 1973; McCann and Roberts 1991; Carpita and Gibeaut 1993). Likewise,
many aspects of the biochemistry of the wall were developed, culminating in the
characterization of several glycosyl hydrolases (e.g. Fry 2004) that correlate with
physiological events in plants, which are controlled by plant hormones and envi-
ronmental cues. With the advance of cellular and molecular biological techniques,
part of the wall research community invested in finding and correlating wall struc-
ture with genes associated with its metabolism (e.g. Holland et al. 2000; Lima et
al. 2001) and also succeeded to find probes (enzymes and monoclonal antibodies)
capable to localise wall components in the cell (e.g. Willats et al. 2001; Pattathil
etal. 2010).
McCann and Roberts (1991) coined the term architecture of the cell wall in a
reference to the existence of three different independent domains in the wall: pec-
tin, cellulose-hemicellulose and proteins. This view implies that the three domains
might interact and this sole interaction would have emergent properties. Carpita
and Gibeaut (1993) showed that at the level of higher plants, the wall structure has
different polysaccharides (e.g. arabinoxylan, -glucan, xyloglucan and mannan)
that may play similar functions. Thus, the architectural model of the wall
points out that cell wall structure is degenerated and that the domains are chang-
ing during evolution, i.e. monocots and ferns changed their main hemicelluloses
from xyloglucan to arabinoxylan and mannans respectively, with a concomitant
decreased pectin domain (Buckeridge 2006; Sarkar et al. 2009).
In higher plants, cellulose, (14)--D-glucan, is the most abundant biologi-
cal material on earth and in dry wood correspond around of 4050% of dry mass.
Hemicellulose is the second polysaccharide representing 2535% of dry mass
and their composition is a mixture of various polymerised monosaccharides, such
as xylose and arabinose or glucose, mannose and galactose whose proportions
change according to the plant tissue and the species.
The principal hemicelluloses found in nature are xyloglucan, galactomannan
or galactoglucomannan, -glucan and glucuronoarabinoxylan (GAX). Xyloglucan
is a major hemicellulose in the primary walls of flowering plants, all non-mono-
cots and about one-half of monocot species. Some seeds store this polysaccha-
ride in large amounts as a resource for the embryo after germination (Buckeridge
et al. 2000a). Xyloglucans typically comprise (14)--D-glucan back-
bones, with three of every four glucose residues substituted with (16)--D-
xylopyranosides, and some of the xylose residues are further substituted with
(1 2)--D-galactopyranosides. The most widespread technique used for stud-
ies on xyloglucan has been the digestion with a Trichoderma endo--glucanase
(Anderson and Stone 1975; Tin et al. 2006). This enzyme hydrolyses xyloglucan
polymers only on unbranched glucosyl residues.
The cell walls from grasses (Poales), which are the main plants used in bio-
energy, have a distinct architecture when compared with non-monocots and non-
Commelinid monocots (Carpita and Gibeaut 1993). Besides the fact that low
levels of pectin are present in their walls, grasses contain as their main hemicel-
luloses the mixed-linked (13), (14)--d-glucan (-glucan) and glucurono-
arabinoxylan (GAX).
-Glucan interlaced by GAX of low degree of substitution and glucomannans,
is thought to be tightly associated with the cellulose microfibrils. GAX of higher
degree of arabinosyl substitution and some glucomannan constitutes the major
pore-determining interstitial material between the microfibrils (see below).
One of the principal enzymes involved in the depolymerization of the
endosperm -glucan is the (13), (14)--D-glucan endohydrolase, whose
activity parallels that of the B. subtilis enzyme in that hydrolysis is restricted to
(1 4)--linked glucosyl residues adjacent to (13)--linked residues on
the non-reducing side (Parrish et al. 1960). Like the B. subtilis endohydrolase,
hydrolysis of -glucan with these enzymes yields mostly cellobiosyl- and cello-
triosyl-(13)-glucosyl oligosaccharides and smaller amounts of larger cellodex-
trin-(13)-glucose oligomers (Anderson and Stone 1975).
In GAX, the arabinose residues are attached at the O-3 positions along the
(14)--linked xylan backbone, and the glucuronic acids are attached to the O-2
positions (Carpita 1996). Arabinoxylans are widespread in the walls of all flower-
ing plants, but in nongramineous species the polymer is of much lower abundance,
and the arabinose residues are attached mostly at the O-2 rather than the O-3 of
the xylosyl units.
A highly substituted GAX (HS-GAX), with six of seven xylosyl branched units
is associated with the maximum growth rate of coleoptiles of the grass species
Lolium multiflorum and a sequence-dependent xylanase was found that requires
branching with GlcA to cleave the neighboring (14)--D-xylosyl linkage
(Nishitani and Nevins 1991). When maize GAXs are depleted of arabinosyl units
by mild-acid hydrolysis, this endo--D-xylanase releases a homogeneous group
of deca- or undecamers of glucuronoxylan. The xylosyl units are also substituted
with acetyl groups at the positions O-2 and O-3 and some methyl groups have
been detected (Carpita 1996).
Porosity of the GAX domain could be determined by the extent of removal
of the appendant units. Some highly substituted GAX remain intercalated in the
small amount of pectins that are also found in the primary wall. Because -glucans
are thought to be tightly associated with individual cellulose microfibrils (Carpita
et al. 2001), GAX might be the main molecule in the interstitial space. However,
Buckeridge and Crivellari (unpublished) have recently performed experiments
with sugarcane walls that did not support this hypothesis. When isolated walls
(alcohol insoluble residues) of leaves and culms of sugarcane were treated with
lichenase (a glycosyl hydrolase that attacks specifically -glucan), most of the pol-
ymer present in the walls was hydrolyzed and, at the same time, its retrieval did
not interfere in the action of endo -xylanase on the wall. Also, De Souza et al.
(2013) observed that -glucans in sugarcane walls are lightly associated with cel-
lulose and/or the cellulose-hemicellulose matrix in the wall. These data suggests
that in sugarcane, although -glucan is present in some association with other pol-
ymers in the wall, they seem to be free for enzyme hydrolysis whereas GAXs are
likely to be strongly bound to other polymers, probably cellulose.
Another distinctive feature of Poales and their relatives is the enrichment of
aromatic substances in nonlignified walls (Carpita 1996). A large portion of the
aromatic substances is esters of hydroxycinnamates (Harris and Hartley 1980).
The GAXs are cross-linked in walls by both esterified and etherified hydroxycin-
namates and by other phenolic substances, as ferulic acid (Iiyama et al. 1994).
On the basis of the knowledge of the structure of the cell wall polysaccharides,
the mode of action of the glycosyl hydrolases and also the genes that encode for
these enzymes, it is now possible to design routes that can deal with the complex
hydrolysis mechanisms of either the less complex seed polysaccharide degrad-
ing systems or even to access more complex wall systems such as the ones from
grasses. The latter is not an easy task as relatively little knowledge exists about the
composition, structure and architecture of the wall of most bioenergy feedstocks.
The cell walls of one of the most important bioethanol feedstocks, sugar-
cane, has been recently analysed by De Souza et al. (2013), who demonstrated
that hemicelluloses are quantitatively the main components of the cell walls.
Regarding carbohydrates alone, whereas hemicelluloses account for ca. 60% of
the walls, cellulose makes around 30%, and pectins about 10%. Hemicelluloses
are of three types: arabinoxylan, -glucan and xyloglucan, the former being about
50% of the hemicelluloses with the other two types sharing the rest in equal
proportion. These authors highlighted the elevated complexity of cell walls and
proposed a model for enzyme hydrolysis in which phenylpropanoids in the wall
would have to be retrieved first in order to open the way for carbohydrate specific
enzymes, pectins being the first to be attacked, followed by the members of the
cellulose-hemicellulose matrix.
In summary, cell wall hydrolysis for bioenergy production purposes can be
thought of as a procedure capable to disassemble cell wall architecture so that
individual polysaccharides would be available to hydrolases. The goal of the pro-
cess is the production of fermentable monosaccharides. On the other hand, even if
the cell wall architecture can be dismantled, the polysaccharides released would
still offer the challenge related to their branching patterns, which are different for
every polysaccharide, resulting in a rather large diversity of combinations that are
named fine structures.
Although the structure of cell wall polymers have been studied for several spe-
cies, less attention is given to fine structure studies. Since the fine structure of
polysaccharides could involve a formation of a glycomic code and interfere in
hydrolysis efficiency (De Souza and Buckeridge, unpublished), the understanding
of branching patterns and its combinations could be an important issue for 2nd
generation processes (see topic 13.4).
In order to illustrate how natural systems could be helpful in bioenergy sec-
tor, in the following sections we will discuss five major systems existent in nature,
which include cell walls modifications by enzymes produced for specific situa-
tions. It must be highlighted here that these items do not intend to review every
biological process exhaustively, but only give key information that are relevant
for the mechanisms of cell wall modifications (including hydrolysis) that could be
used to transform plants and make them more suitable as bioenergy crops.
13.3Cell Wall Degradation During Plant-Microorganism

Interaction
When a pathogen invades a plant tissue, a biochemical conflict takes place, in

which host and invader will produce a series of substances, the former to defend
itself and the latter to penetrate in the cells (Garcia-Brugger et al. 2006). Among
the strategies used by both organisms there are cell wall hydrolases (Walton 1994).
In this way, the process of plant-microorganism interaction offers an opportunity
to find ways to efficiently disassemble the plant cell wall. Both, saprophytic and
pathogenic microorganisms produce extracellular enzymes that can degrade plant
polysaccharides. These microorganisms, including bacteria and fungi, and also
invertebrates as nematodes can digest cell wall polymers in order to obtain sug-
ars and energy for their growth as well as to penetrate and colonize the cells in a
plant tissue (Walton 1994; Annis and Goodwin 1997). In most cases, plants will
respond with a reaction named hypersensitivity, which includes programmed cell
death and with that can restrict the infectious region to a small area surrounding
the infection. However, in these cases, pathogens can use this opportunity to feed
on dead cells and gain nutrients and with that they may become even more inva-
sive (Greenberg and Yao 2004).
The first class of cell wall hydrolases produced by the pathogen is the pecti-
nases (Tomassini et al. 2009). They are thought to loosen the tissue through diges-
tion of the middle lamella and consequently kill the plant cells that they attack
(Cooper 1983). The pectinases known to be involved in fungal and bacterial inva-
sion are endo- and exo-polygalacturonases, pectate-lyases, and pectin methylester-
ases (Walton 1994; Lagaert et al. 2009). It is also believed that even saprophytes
need to use cell wall degradation strategies in order to establish interaction with
the plant and for that they also produce pectinases (e.g. Marques et al. 2006).
Pectins are polysaccharides that control cell wall porosity. Some studies dem-
onstrated that the porosity of the substrate walls is the main limiting factor in the
enzymatic hydrolysis of lignocellulosic biomass (Chandra et al. 2007). Thus, the
knowledge about how microorganisms use the pectinases to invade the plant could
be a way to improve the biomass hydrolysis to 2G bioethanol.
Futhermore, several other enzymes that are able to hydrolyse cell walls can
be produced by microorganisms. Recently, the need for enzymes to be used in
the production of bioethanol from biomass led many groups to produce a large
amount of scientific information about glycosyl hydrolases from microorgan-
isms. The isolation and characterization of enzymes as well as cloning their genes,
open a way to engineering those derived from microorganisms, i.e. modify them
to perform hydrolysis more efficiently (Serpa and Polikarpov 2011; Ward 2011).
Then, new cocktails of enzymes could be designed that would efficiently produce
free fermentable sugars for bioethanol production. A review of the enzymes from
microorganisms used for bioenergy has been recently published (Polizeli et al.
2011).
13.4Seed Cell Wall Storage Mobilization
Cell Wall Storage Polysaccharides (CWSP) are found as the principal storage
compound in seeds of many taxonomically important groups of plants. These
groups developed extremely efficient biochemical mechanisms to disassemble cell
walls and use the products of hydrolysis for growth. During evolution, several dif-
ferent groups of polymers have been selected as reserves (e.g. mannans, galacto-
mannans, galactoglucomannans, arabinogalactans and xyloglucans) (Buckeridge
et al. 2000a, 2000b; Buckeridge 2010). The selective pressures that ended up pro-
ducing CWSP were directed towards an increase their proportion by introducing
a storage cell wall between already existent primary walls (Tin, Braga, Pattathil,
Hahn and Buckeridge, unpublished).
The CWSP have different chemical structures that are related with different
biological functions. Mannans and galacto- and glucomannans that are present in
palm and coffee seeds and lettuce and tomato respectively are insoluble in water
and display strong intermolecular interaction. Because of these characteristics,

their biological function is usually associated with conferring hardness to plant tis-
sues. Endo--mannanase is the principal enzyme involved in mannan hydrolysis
in all species studied and several genes have been cloned whose expression is spe-
cific to the seed (Bewley et al. 1997; Lisboa et al. 2006; Gong and Bewley 2007).
In legumes, the main function of the endospermic cell walls appear to be stor-
age, with the yield of galactomannan reaching more than 30% of the seed dry
weight in many species (Buckeridge et al. 2000a). These walls are thickened
with galactomannan and in certain cases (e.g. Trigonella foenum-graecum and
Schyzolobium parayba) the protoplasm disappears, giving place to the storage
wall. In these cases, the endosperm is non-living and degradation is performed by
three enzymes (-galactosidase, endo--mannanase and exo--mannosidase) made
in the aleurone layer (Reid 1971; Buckeridge and Dietrich 1996).
Differently from being stored in the endosperm, arabinogalactan and xyloglu-
cans are stored in cotyledons. This is extremely relevant because the cotyledon is,
evolutionarily speaking, an adapted leaf and the integration of the metabolism of
these polysaccharides with the entire plant is likely to preserve several features of
the leaf cell walls.
Pectin polymers are also found as cell wall storage polysaccharides, notably
arabinogalactan (Buckeridge et al. 2000b). The cotyledons of lupins (especially
Lupinus angustifolius) have been used as a model to study mobilization of this
polymer.
Lupinus angustifoluis seeds accumulate high proportions of galactan
(Buckeridge and Reid 1994). These authors showed that an exo--(1,4)-
galactanase purified from L. angustifolius seed acts specifically on this type
of galactan. This enzyme is analogous to galactanases found in fruits (TBG4 in
tomato for examplesee below) where it is thought to play a major function in
the changes of porosity in cell walls (Brummell 2006).
Of the CWSP known, xyloglucan seems to be the more complex in most
senses. Its basic structure is similar to the primary wall xyloglucans. They have
a backbone composed of (14)--linked glucan (like cellulose) with regular
branching with (16)--linked xylosyl residues that can be branched further
with (12)--linked galactosyl residues. There is no case reported where fucose
was detected in storage xyloglucan. Four enzymes responsible for storage xylo-
glucan degradation have been detected, purified and characterized (-xylosidase,
-galactosidase, -glucosidase and xyloglucan-endo--glucanase, latter on
renamed xyloglucan transglucosylase/hydrolase (XTH) (Buckeridge et al. 2000a;
Buckeridge 2010).
In Hymenaea, Buckeridge et al. (1997) discovered that storage xyloglucan con-
tains two types of constitutive blocks in the main chain, i.e., XXXG and XXXXG.
Later on, Tin et al. (2006) showed that these blocks are not combined randomly
in the polymer molecules, which led to the proposition that storage xyloglucans
might have a combination of branching patterns that forms a code that have to be
decrypted by the enzymatic system in order to be hydrolysed (Buckeridge 2010).
Indeed, the fact that branching of cell wall polymers might not be random has
been also observed early on for galactomannans (Reid and Edwards 1995).
As this kind of information is not available in most cases, research should be

urgently directed to that in order to complement the efforts in the search for new
enzymes as well as enzyme engineering.
13.5Fruit Ripening
During ripening, the cells of fruits display changes in cell walls so that the final
result will afford the release of seeds for dispersion. Several species have been
adapted to human consumption and became sources of food, being nowadays
extremely important in agriculture. That is the case of tomato, apple, pear, melon,
papaya, grape to name but a few. Due to the economic importance of fruits,
development and ripening have been intensively studied, with tomato being one
of the most studied fruit regarding studies of cell wall changes related to texture
(Brummell and Harpster 2001; Brummell 2006). Changes in fruit texture that
occur during ripening in many cases lead to softening, this process being slightly
different for each species (Harker et al. 1997; Brummell 2006). Nonetheless, in
all cases the texture is related to changes in cell walls that are associated with the
production of several hydrolases. Due to the fact that enzymes can act to hydro-
lyse fruit cell wall polysaccharides and/or to allow structural changes that modify
the mechanical properties of plant tissues, fruit systems can be considered as an
opportunity to find mechanisms of cell wall degradation in plants that could help
developing biotechnology for bioenergy.
In fruit systems, the most studied degradation process is the activity of endopo-
lygalacturonases, pectin methylesterases and -galactosidases on pectic matrix of
the walls. Together with the detection of enzyme activities, changes in cell walls
of tomato are consistent with a large decrease in uronic acid as well as even larger
decreases in galactose in wall fractions (Ahmed and Labavitch 1980; Rose et al.
1998; Brummell and Harpster 2001; Brummell 2006).
It has been demonstrated that pectinases are associated with ripening and sof-
tening through degradation of endopolygalacturonases present in the middle
lamella. The action of this enzyme is thought to lead to cell separation, a phenom-
enon that is linked to the subsequent softening of the fruits and change in their
texture and taste (Fray and Grierson 1993; Barsan et al. 2010).
The xyloglucan-cellulose network has also been pointed out as an important
target in ripening (Rose and Bennett 1999; Brummell 2006). These authors high-
light the role of expansions and XTHs in the process and the former authors even
compare the process of fruit ripening with the cell expansion system in hypoco-
tyls. Brummell (2006) reviewed fruit ripening from the viewpoint of the wall
changes, putting forward the idea that whereas pectin degradation varies among
different fruit species (Brummell 2006), the xyloglucan-cellulose network seems
to undertake changes that are much more consistent among species. However,
the author also highlights the fact that, contrarily to what has been observed for
pectins, xyloglucan small fragments have not been detected. Thus, the effects
of expansin and enzymes on the xyloglucan-cellulose matrix seem to be more
towards a relaxation of this matrix rather than hydrolysis. This seems to lead to
cell wall swelling that happens concomitantly with pectin solubilization.
The increase in pore sizes, thought to be related to the attack of exo--
galactanase in pectin matrix (Brummell 2006), probably opens the way for expan-
sions, XTHs, cellulases, as well as other debranching enzymes, that will attack
the cellulose-hemicellulose matrix of the wall, finally changing wall texture as a
whole. Indeed, according to Carpita et al. (1979), pore sizes of the walls in plant
tissues would be in the range of 3540, whereas the glycosyl hydrolases stoke
radius [e.g. a 20kDa protein would be approximately 2.5nm according to Carroll
and Somerville (2009)] is in general higher than the pore sizes, so that this is a
clear limitation for penetration of enzymes in the cell wall matrix.
Although endo--glucanases have been found to increase during fruit ripen-
ing, very little evidence has been produced that cellulose would be hydrolysed.
However, it has been speculated that cellulases would be capable to act on the sur-
face of microfibrils (Brummell 2006) without, however, any assignment of a func-
tion for this event.
In spite of the fact of hydrolysis, in a broad sense, being the main process in
course during fruit ripening, other steps such as cell separation and expansion,
have been observed (Brummell 2006). On the other hand, differently from some
other cell wall degradation processes such as aerenchyma formation and abscis-
sion (see below), fruit ripening seems do not include programmed cell death.
From a general point of view, fruit ripening is a process in which walls are
firstly attacked by pectinases that increases wall porosity, probably allowing the
action of other enzymes like XTHs and endo--glucanases that will transform
the cellulose-hemicellulose matrix, promoting swelling and increasing even more
wall porosity. The extensins that bind to the cellulose-hemicellulose matrix of the
wall also could participate of these processes, being probably the first proteins to
interact with the wall matrix. Thus, the cell wall alteration plus a higher hydration
capacity given by the presence of pectins with lower molecular weight modifies
fruits softening.
All this knowledge is extremely useful for bioenergy purposes, as it makes pos-
sible to understand details of how plants can manipulate their own walls. These
events could be compared with the ones that occur during biomass pretreatment
process used to achieve the 2G bioethanol. The possibility of applying such
knowledge in bioenergy crops to produce similar effects in grass stems cell walls
(e.g.) could lead to the establishment of a biological pretreatment that could be
implemented using synthetic biology tools in the future.
13.6Abscission
Abscission is a general term applied to the processes that takes place during the
detachment of leaves, fruits, seeds and also cotyledons, lateral branches and many
reproductive structures. In all cases studied, extremely precise biochemical path-
ways lead to cell wall modification and breakdown in order to form a fracture line
known as the abscission zone. The events that occur during abscission and dehis-
cence processes are cleary related to cell wall modification, in a very similar way
to the other ones reported in this chapter.
Roberts et al. (2002), highlight the facts that endopolygalacturonases and endo-
-glucanases are closely related to the processes of cell separation. In fruits like
tomato, silencing of abscission-related polygalacturonases (e.g. TAPG1) led to a
delay in abscission and increased the force needed to break the abscission zone in
explants treated with ethylene (Jiang et al. 2008).
In the abscission zone, cortex cells of different plant tissues are thought to be
positionally differentiated target cells that will receive the signal from hormones
(ethylene, auxin, abscisic acid, gibberellin) in such a balance that the cross talk
among these will lead to transdifferentiation and finally to abscission. McManus
et al. (1998) demonstrated the existence of transdifferentiation in mature corti-
cal cells of bean abscission tissues. According these authors, this means that cells
from the cortex of some plant tissues have the flexibility to differentiate into func-
tionally competent ethylene-responsive cells that exhibit a gene expression com-
patible with an abscission cell.
Although not reported in most cases, cell wall degradation is thought to
occur during abscission, since detection of activity of endo--glucanase has
been reported for many abscission systems (Roberts et al. 2002). Also, accu-
mulation of mRNA related with genes encoding Cel1 and Cel4 in tomato fruit
and flower abscission zones have been reported (Gonzles-Carranza et al.
1998).
The similarity of this process with the ones described above is remarkable.
Abscission is widespread in plants, occurring in many different tissues of different
organs in plants. It includes wall separation and expansion, which are associated
with pectin degradation as well as modifications in the cellulose hemicellulose
matrix. However, programmed cell death has not been reported for abscission pro-
cesses, according to the literature revised for this chapter. Thus, there seems to be
a great variety of combinations of action of hydrolases to perform cell separation
and hydrolysis so that it could be used to design technologies associated to 2G
bioethanol production.
13.7Aerenchyma Formation
Aerenchyma comprises a series of interconnected gas chambers developing on

parenchymatic tissue (cortex) of shoots and roots as a result of sequencial events,
showing typical features of a programmed cell death process. Its formation is
regarded as a response to oxygen shortage (Gunawardena et al. 2001a), exog-
enous ethylene (Gunawardena et al. 2001b), nitrogen, phosphorus and/or sulphur
starvation (He et al. 1994; Siyiannis et al. 2011), mechanical impedance (He et
al. 1996b), oxidative stress (Steffens et al. 2011) and osmotic stress (Karahara
etal.2012).
In opposition to the inducible aerenchyma, the constitutive formation is com-

monly observed in aquatic species as Juncus effusus (Visser and Bgemann 2006)
and Sagittaria lancifolia (Schussler and Longstreth 2000), and also in maize rela-
tives (Mano et al. 2007) and in rice wetland species (Justin and Armstrong 1991),
independently of environmental stimuli. In both cases, its development lies upon a
cell separation process termed schizogeny (thought of as a cell separation process
that does not include cell wall hydrolysis). Alternatively, for the lysogenic process
(i.e. cell separation followed by cell wall hydrolysis) to take place, the cells tar-
geted to form aerenchyma clearly undergo a programmed cell death. It is essential
to mention that both processes rely on polysaccharide hydrolysis, whereas the cell
separation in schizogenous aerenchyma formation depends on middle lamella deg-
radation. As for the lysogenous process, cell wall modifications occur and for that
a wide array of enzymes would be required.
Concerning cell wall modifications, most of the aerenchyma-related data is
restricted to the lysigenic and inducible formation in maize roots. As reported for
other cell degradation events, in this model the inductive role is played by ethylene
(He et al. 1992, 1996a; Gunawardena et al. 2001a). One of the earliest signs of
aerenchyma formation is the ethylene accumulation, followed by cell death and
endo--glucanases activity (Kawase 1979; He et al. 1996a). Indeed, inhibitors of
cell death or ethylene biosynthesis block the enhancement of endo--glucanases
activity (He et al. 1994, 1996a), suggesting the coupling between cell wall degra-
dation, ethylene signalling and programmed cell death.
Another potentially degradative enzyme associated with aerenchyma forma-
tion is endo--xylanase. The rising activity levels of this enzyme were obtained
in waterlogged maize roots, coinciding with the rising of ethylene levels (Bragina
et al. 2001). Also concerning ethylene induction, the role of XTH, a putative cell
wall loosening enzyme, was evaluated during aerenchyma formation. Transcripts
of certain XTH accumulated in flooded maize roots forming aerenchyma, reaching
its maximum within 12h and remaining high for 144h during flooding. On the
other hand, the inhibition of ethylene synthesis blocked any rise of this transcript
accumulation, while exogenous ethylene led to XTH transcript accumulation even
under aerobic conditions (Saab and Sachs 1996). A transcriptional analysis cor-
roborated the induction of XTH expression during the aerenchyma formation in
maize roots, as well as for endo--glucanase (Rahji et al. 2011).
Even though the enzymatic or transcriptional approaches have been used to
suggest the link between a set of glycosil hydrolases and aerenchyma formation,
to our knowledge, data regarding cell wall modifications in situ have been reported
only by Gunawardena et al. (2001b), who documented the distinctive levels of
methyl esterification changing in the middle lamella in cross sections of the cortex
of the maize root along aerenchyma formation. These authors observed that those
changes initiated before ultrastructural modifications typical of cells undergoing
programmed cell death that have been reported in a previous work (Gunawardena
et al. 2001a). Indeed, rise in the expression of pectin methyl esterases, pectate
lyase and an endopolygalacturonase have been observed during aerenchyma
formation (Rahji et al. 2011). These data not only corroborate the existence of
homogalacturonan modifications but also suggest that more severe changes on the
cell wall structure and not only on the middle lamella can occur. We have observed
a similar pattern of events in sugarcane roots, not only with the pectinases, but also
with the observation of the rise in gene expression of several genes that encode
hemicellulases (e.g. lichenase, endo--xylanase, XTH, -glucosidaseunpub-
lished results).
Gunawardena et al. (2001a) reported the formation of vesicles and vacuole rup-
ture while the cell wall appeared intact. It is reasonable to hypothesize that the
release of hydrolytic enzymes in this step could be responsible for the cell wall
degradation (Bouranis et al. 2007).
The data produced that are related with aerenchyma formation to date indicates
that this phenomenon is similar to the other processes discussed above in the sense
that cell walls are modified, including cell separation, hemicellulose-cellulose
matrix relaxation and perhaps some hydrolysis of cellulose. Thus, the knowlegde
about aerenchyma formation can also be considered as a possible target for use
in biotechnology that includes cell wall modifications for bioenergy purposes.
Recently we have produced a consistent set of results describing events during the
aerenchyma formation in sugarcane roots. We found that the events are essencially
similar to maize (e.g.), with the exception that in sugarcane it is constitutively acti-
vated. Our results (manuscripts in preparation) show very clearly that cell sepa-
ration, programmed cell death, hemicellulose-cellulose changes and some attack
to cellulose occurs in sugarcane. We are now searching for the transcription fac-
tors responsible for regulating this processes aiming at possibly inducing them in
culms so that this process might be used as a biological pretreatment in the future.
13.8Cell Wall Modifications in Natural Systems in the

Context of Bioenergy
As mentioned above, in order to establish 2G bioethanol production, the retrieval

of energy from wall polymers would have to include cell wall modification. From
the different systems discussed along this chapter, it is possible to think of com-
bining some of them in order to help increasing the production of bioethanol from
bioenergy feedstocks that are being studied for use in 2G processes.
In Table13.1 the different modules that plants apparently use to transform their
cell walls in order to perform different biological functions are shown in perspec-
tive. Six events seem to be common to most of the natural processes that clearly
involve modifications in cell walls. Some of these events have been investigated
more deeply and others are implied, but not proven, as no direct experimental evi-
dences have been produced for them yet. In Table13.1, the events are placed in
an order (from left to right) that they seem to occur in all systems. However, some
steps are missing in some of the processes (e.g. cell separation in storage cell wall
mobilization and programmed cell death in fruit ripening).
Table13.1Correlation among situations that include cell wall modifications and natural events
in plant biology where these situations are key to biological function
Target Cell Cell Programmed Hemicellulose Cellulose
Processes cells separation expansion cell death hydrolysis hydrolysis
Plant ? YES ? YES YES YES
microorganism
interaction
Storage cell wall ? NO YES ? YES ?
mobilization
Fruit ripening YES YES YES NO YES YES/NO
Abscission YES YES YES ? YES YES
Aerenchyma YES YES YES YES YES ?
formation
YES=there is literature showing that it occurs in most (or all) systems studied; NO=not
observed in the systems studied to date; YES/NO=observed only in some of the systems stud-
ied to date; ?=not reported to occur to date
The combination of events is consistent with a chain of processing steps that

would follow a sequence of modifications in the walls that are apparently similar
in the five processes reviewed in this chapter. The general idea is that in a par-
enchymatic tissue, a given group of cells would be targeted to start the process
(as seen in abscission according to Roberts et al. (2002) and also in aerenchyma
formation). The following step would be cell separation that probably involves
the action of endo-polygalacturonase(s) on the pectins of the middle lamella. This
might be involved with the release of calcium that could be responsible for the
signaling related to programmed cell death. The latter seems to occur in parallel
with cell expansion and subsequent production and release glycosyl hydrolases
that would attack hemicelluloses and cellulose. Every one of these events can be
regarded as a module that can or not be used in the complete process that leads to
partial or complete cell wall modification.
There are two possible ways to use this information for biotechnology pur-
poses: (1) using them for production of transformed plants; or (2) express some
of the hydrolase genes heterologously in order to use them as additions to enzyme
cocktails already in use in industry.
For the first option, it should be possible to use the modern tools of synthetic
biology in order to induce cell separation and expansion along with some modi-
fication of the hemicellulose-cellulose matrix. In this case, plants would produce
biomass with the characteristics expected to be analogous to a pretreated biomass
material, i.e. materials that would be easily hydrolyzed when being processed
in industry, especially during the hydrolysis process. For advancing the second
option, some of the enzymes known to be present during hydrolysis processes in
bioenergy feedstocks, such as sugarcane, miscanthus, maize, poplar, willow and
others should be studied regarding their specificity. Some of these enzymes may
be heterologously expressed for production and added to existent cocktails.
In both cases, there is a relatively long way to go in terms of production of the

proof of concepts that will be necessary in order to develop the necessary tech-
nologies. However, in some cases, such processes are already present in plants and
could be improved using transformation, possibly of transcription factors that trig-
ger the whole process.
It is important to note here that wood feedstocks would probably have to be
treated differently regarding the strategies discussed in this chapter. Wood feed-
stocks have large proportions of dead cells composed of high proportions of cellu-
lose and also with a much higher degree of lignification. For this type of bioenergy
feedstock, an addition to the strategy would be the modification of lignin com-
position to facilitate access to the cell wall polysaccharides, which in fact is cur-
rently under investigation (Vanholme et al. 2013). As grasses can have some lignin
and phenypropanoids in their cell walls, as well as the cells of the vascular sys-
tem, whose walls bear many features in common with the wood tissues, the use
of the strategies proposed in this chapter for tissues containing large proportion
or primary cell walls, along with the ones being developed for decreasing lignin
interference, would probably have to be coupled in the future in order to improve
bioenergy production even more.
Based on the examples discussed in this chapter, it is very likely that natural
processes that involve cell wall modifications in plants will be useful as additions
to 2G bioethanol technologies, via the use of modern molecular techniques associ-
ated to systems and synthetic biology.
Acknowledgments This work is part of the production of the Instituto Nacional de Cincia e
Tecnologia do Bioetanol-INCT do Bioetanol (FAPESP 2008/57908-6 and CNPq 574002/2008-1)
and of the Centro de Processos Biolgicos e Industriais para Biocombustveis-CeProBIO (FAPESP
2009/52840-7 and CNPq 490022/2009-0). Financial support by FAPESP Projects 2010/17104-5,
2010/17070-3, 2011/07586-5 and 2011/02344-3.
References
Ahmed AE, Labavitch JM (1980) Cell wall metabolism in ripening fruit 1. Cell wall changes in
ripening Bartlett pears. Plant Physiol 65:10091013
Anderson MA, Stone BA (1975) A new substrate for investigating the specificity of glucan
hydrolases. FEBS Lett 52:202207
Annis SL, Goodwin PH (1997) Recent advances in the molecular genetics of plant cell wall-
degrading enzymes produced by plant pathogenic fungi. Eur J Plant Pathol 103:114
Balat M (2011) Production of bioethanol from lignocellulosic materials via the biochemical path-
way: a review. Energy Converse Manag 52:858875
Barsan C, Sanchez-Bel P, Rombaldi C, Egea I, Rossignol M, Kuntz M et al (2010)
Characteristics of the tomato chromoplast revealed by proteomic analysis. J Exp Bot
61:24132431
Bewley JD, Burton RA, Morohashi U, Fincher GB (1997) Molecular cloning of a cDNA encod-
ing a (14)--mannan endohydrolase from the seeds of germinated tomato (Lycopersicon
esculentum). Planta 203:454459
Bouranis DL, Chorianopoulou SN, Siyiannis VF, Protonotarios VE, Hawkesford MJ (2007)
Lysigenous aerenchyma development in roots: triggers and cross-talks for a cell elimination
program. Int J Plant Dev Biol 1:127140
Bragina TV, Martinovich LI, Rodionova NA, Bezborodov AM, Grineva GM (2001) Ethylene-
induced activation of xylanase in adventitious roots of maize as a response to the stress effect
of root submersion. Appl Biochem Microbiol 37:722725
Brummell DA (2006) Cell wall disassembly in ripening fruit. Funct Plant Biol 33:103119
Brummell DA, Harpster MH (2001) Cell wall metabolism in fruit softening and quality and its
manipulation in transgenic plants. Plant Mol Biol 47:311340
Buckeridge MS (2006) Implications of emergence, degeneracy and redundancy for the modeling
of the plant cell wall. In: Hayashi T (ed) The science and the lore of the plant cell wall: bio-
synthesis, structure and function. BrownWalker Press, Boca Raton, pp 4147
Buckeridge MS (2010) Seed cell wall storage polysaccharides: models to understand cell wall
biosynthesis and degradation. Plant Physiol 154:17
Buckeridge MS, De Souza AP (2013) The fine structure of cell wall polysaccharides might be
related to the existence of a glycomic code in plant cell walls. In: Lam E (ed) Compendium
of bioenergy plants: sugarcane (submitted)
Buckeridge MS, Dietrich SMC (1996) Mobilisation of the raffinose family oligosaccharides
and galactomannan in germinating seeds of Sesbania marginata Benth (Leguminosae-
Faboideae). Plant Sci 117:3343
Buckeridge MS, Reid JSG (1994) Purification and properties of a novel -galactosidase or exo-
(14)--D-galactanase from the cotyledons of germinated Lupinus angustifolius L. seeds.
Planta 192:502511
Buckeridge MS, Crombie HJ, Mendes CJM, Reid JSG, Gidley MJ, Vieira CJ (1997) A new fam-
ily of xyloglucan oligosaccharides from cotyledons of Hymenaea courbaril: structure deter-
mination of the oligosaccharide XXXXG by enzymatic sequencing and NMR. Carbohydr
Res 303:233237
Buckeridge MS, Dietrich SMC, Lima DU (2000a) Galactomannans as the reserve carbohy-
drate of legume seeds. In: Gupta AK, Kaur N (eds) Developments in crop science. Elsevier
Science BV, Amsterdam, pp 283316
Buckeridge MS, Santos HP, Tin MAS (2000b) Mobilisation of storage cell wall polysaccharides
in seeds. Plant Physiol Biochem 38:141156
Buckeridge MS, De Souza AP, Arundale RA, Anderson-Teixeira KJ, DeLucia E (2012) Ethanol
from sugarcane in Brazil: a midway strategy for increasing ethanol production while maxi-
mizing environmental benefits. GCB Bioenergy 4:119126
Carpita NC (1996) Structure and biogenesis of the cell walls of grasses. Ann Rev Plant Physiol
Plant Mol Biol 47:445476
Carpita NC, Gibeaut DM (1993) Structural models of primary cell walls in flowering plants: con-
sistency of molecular structure with the physical properties of the walls during growth. Plant
J 3:130
Carpita NC, Sabularse D, Montezinos D, Delmer DP (1979) Determination of the pore size of
cell walls of living plant cells. Science 205:11441147
Carpita N, Tierney M, Campbell M (2001) Molecular biology of the plant cell wall: searching for
the genes that define structure, architecture and dynamics. Plant Mol Biol 47:15
Carroll A, Somerville C (2009) Cellulosic biofuels. Ann Rev Plant Biol 60:165182
Chandra RP, Bura R, Mabee WE, Berlin A, Pan X, Saddler JN (2007) Substrate pretreatment:
the key to effective enzymatic hydrolysis of lignocellulosics? Biofuels Adv Biochem Eng
Biotechnol 108:6793
Cooper RM (1983) The mechanisms and significance of enzymatic degradation of host cell walls
by parasites. In: Callow JA (ed) Biochemical plant pathology. Wiley, Chichester, pp 101137
De Souza AP, Leite DCC, Pattathil S, Hahn MG, Buckeridge MS (2013) Composition and struc-
ture of sugarcane cell wall polysaccharides: implications for second-generation bioethanol
production. Bioenerg Res 6:564579
Dos Santos WD, Gmez EO, Buceridge MS (2011) Bioenergy and the sustainable revolution.
In: Buckeridge MS, Goldman GH (eds) Routes to cellulosic ethanol. Springer, New York, pp
1526
Fray RG, Grierson D (1993) Molecular genetics of tomato fruit ripening. Trends Genet
9:438443
Fry S (2004) Primary cell wall metabolism: tracking the careers of wall polymers in living plant
cells. New Phytol 161:641675
Garcia-Brugger A, Lamotte O, Vandelle E, Bourque S, Lecourieux D et al (2006) Early signaling
events induced by elicitors of plant defenses. Mol Plant Microbe Interact 19:711724
Gmez LD, Steele-King CG, McQueen-Mason SJ (2008) Sustainable liquid biofuels from bio-
mass: the writings on the walls. New Phytol 178:473485
Gong X, Bewley JD (2007) Shorting out the LeMANs: endo--mannanase genes and their
encoded proteins in tomato. Seed Sci Tech 17:143154
Gonzles-Carranza ZH, Lozoya-Gloria E, Roberts JA (1998) Recent developments in abscission:
shedding light on the shedding process. Trends Plant Sci 3:1014
Greenberg JT, Yao N (2004) The role and regulation of programmed cell death in plant-pathogen
interactions. Cell Microbiol 6:201211
Gunawardena A, Lan H, Pearce DM, Jackson MB, Hawes CR, Evans DE (2001a)
Characterization of programmed cell death during aerenchyma formation induced by ethyl-
ene or hypoxia in roots of maize (Zea mays L.). Planta 212:205214
Gunawardena A, Pearce DM, Jackson MB, Hawes CR, Evans DE (2001b) Rapid changes in cell
wall pectic polysaccharides are closely associated with early stages of aerenchyma forma-
tion, a spatially localized form of programmed cell death in roots of maize (Zea mays L.)
promoted by ethylene. Plant Cell Environ 24:13691375
Harker FR, Redgwell RJ, Hallet IC, Murray SH, Carter G (1997) Texture of fresh fruit. Hortic
Rev 20:121124
Harris PJ, Hartley RD (1980) Phenolic constituents of the cell walls of monocotyledons.
Biochem Syst Ecol 8:153160
He CJ, Morgan P, Drew MC (1992) Enhanced sensitivity to ethylene in nitrogen of phosphate-
starved roots of Zea mays L. during aerenchyma formation. J Plant Physiol 98:137142
He CJ, Drew M, Morgan PW (1994) Induction of enzymes associated with lysigenous aeren-
chyma formation in roots of Zea mays during hypoxia or nitrogen starvation. J Plant Physiol
105:861865
He CJ, Pinlayson SA, Drew MC, Jordan WR, Morgan PW (1996a) Ethylene biosynthesis during
aerenchyma formation in roots of Zea mays subjected to mechanical impedance and hypoxia.
J Plant Physiol 112:16791685
He CJ, Morgan PW, Drew MC (1996b) Transduction of an ethylene signal is required for cell
death and lysis in the root cortex of maize during aerenchyma formation induced by hypoxia.
J Plant Physiol 112:463472
Himmel ME, Adney WS, Baker JO, Elander R, McMillan JD, Nieves RA, Sheehan JJ, Thomas
SR, Vinzant TB, Zhang M (1997) Advanced bioethanol production technologies: a prospec-
tive. In: Woodward J, Saha B (eds) ACS symposium series, vol 666. American Chemical
Society, Washington, DC, pp 245
Himmel ME, Ruth MF, Wyman CE (1999) Cellulase for commodity products from cellulosic
biomass. Curr Opin Biotechnol 10:358364
Himmel ME, Ding SY, Johnson DK, Adney WS, Nimlos MR, Brady JW, Foust TD (2007)
Biomass recalcitrance: engineering plants and enzymes for biofuels production. Science
315:804807
Holland N, Holland D, Helentjaris T, Dhugga KS, Xoconostle-Cazares B, Delmer DP (2000)
A comparative analysis of the plant cellulose synthase (CesA) gene family. Plant Physiol
123:13131323
Iiyama K, Lam TBT, Stone BA (1994) Covalent crosslinks in the cell wall. Plant Physiol
104:315320
Jiang CZ, Lu F, Imsabai W, Meir S, Reid MS (2008) Silencing polygalacturonase expression
inhibits tomato petiole abscission. J Exp Bot 59:973979
Justin SHFW, Armstrong W (1991) Evidence for the involvement of ethene in aerenchyma for-
mation in adventitious roots of rice (Oryza sativa L.). New Phytol 118:4962
Kaida R, Kaku T, Baba K, Oyadomari M, Watanabe T, Nishida K, Kanaya T, Shani Z, Shoseyov

O, Hayashi T (2009) Loosening xyloglucan accelerates the enzymatic degradation of cellu-
lose in wood. Mol Plant 2:904909
Karahara I, Umemura K, Soga Y, Akai Y, Bando T, Ito Y, Tamaoki D, Uesugi K, Abe J, Yamauchi
D, Mineyuki Y (2012) Demonstration of osmotically dependent promotion of aerenchyma
formation at different levels in the primary roots of rice using a sandwich method and X-ray
computed tomography. Ann Bot 110:503509
Kawase M (1979) Role of cellulase in aerenchyma development in sunflower. Amer J Bot
66:183190
Keegstra K, Talmadge K, Baurer WD, Albersheim O (1973) Structure of the plant cell walls.
Lagaert S, Belin T, Volckaert G (2009) Plant cell walls: protecting the barrier from degradation
by microbial enzymes. Stem Cell Dev Biol 20:10641073
Lima DU, Santos HP, Tin MAS, Molle FD, Buckeridge MS (2001) Patterns of expression of
cell wall related genes in sugarcane. Gen Mol Biol 24:191198
Lisboa CGS, Tin MAS, Buckeridge MS (2006) Endo--mannanase from the endosperm of
seeds of Sesbania virgata (Cav.) Pers. (Leguminosae): purification, characterisation and
importance for germination and early seedling growth. Braz J Plant Physiol 18:269280
Mano Y, Omori F, Takamizo T, Kindiger B, Bird R, Loaisiga C, Takahashi H (2007) QTL map-
ping of root aerenchyma formation in seedlings of a maize e rare teosinte Zea nicaraguen-
sis cross. Plant Soil 295:103113
Marques MR, Buckeridge MS, Braga MR, Dietrich SMC (2006) Characterization of an extracel-
lular endopolygalacturonase from saprobe Mucor ramosissimus Samutsevitsch and its action
as trigger of defensive response in tropical plants. Mycopathologia 162:337346
McCann MC, Roberts K (1991) Architecture of the primary cell wall. In: Lloyd CW (ed) The
cytoskeletal basis of plant growth and form. Academic Press, London, pp 109129
McManus MT, Thompson DS, Merriman C, Lyne L, Osborne DI (1998) Transdifferentiation of
mature cortical cells to functional abscission cells in bean. Plant Physiol 116:891899
Nishitani K, Nevins DJ (1991) Glucuronoxylan xylanohydrolase: a unique xylanase with the
requirement for appendant glucuronosyl units. J Biol Chem 266:65396543
Ong RG, Chundawat SPS, Hodge DB, Keskar S, Dale B (2013) Linking plant biology and pre-
treatment: understanding the structure and organization of the plant cell wall and interac-
tions with cellulosic biofuel production. In: Carpita NC, Buckeridge MS, McCann MC (eds)
Plants and bioenergy. Springer, New York
Parrish FW, Perlin AS, Reese ET (1960) Selective enzymolysis of poly--D-glucans, and the
structures of the polymers. Can J Chem 38:20942104
Pattathil S, Avci U, Baldwin D, Swennes AG, McGill JA, Popper Z et al (2010) A comprehensive
toolkit of plant cell wall glycan-directed monoclonal antibodies. Plant Physiol 153:514525
Polizeli MLTM, Corra ECP, Polizeli AM, Jore JA (2011) Hydrolases from microorganisms used
for degradation of plant cell wall and bioenergy. In: Buckeridge MS, Goldman GH (eds)
Routes to cellulosic ethanol. Springer, New York, pp 117136
Rahji I, Yamauchi T, Takahashi H, Nishiuchi S, Shiono K, Watanabe R, Mliki A, Nagamura Y,
Tsutsumi N, Nishizawa NK, Nakazono M (2011) Identification of genes expressed in maize
root cortical cells during lysigenous aerenchyma formation using laser microdissection and
microarray analyses. New Phytol 190:351368
Reid JSG (1971) Reserve carbohydrate metabolism in germinating seeds of Trigonella foenun-
graecum L. (Legum). Planta 100:131142
Reid JSG, Edwards M (1995) Galactomannans and other cell wall storage polysaccharides in
seeds. In: Stephen AM (ed) Food polysaccharides and their applications. Marcel Dekker,
New York, pp 155186
Roberts JA, Elliott KA, Gonzalez-Carranza Z (2002) Abscission, dehiscence, and other cell sepa-
ration processes. Ann Rev Plant Biol 53:131158
Rose JKC, Bennett AB (1999) Cooperative disassembly of the cellulose-xyloglucan network

of plant cell walls: parallels between cell expansion and fruit ripening. Trends Plant Sci
4:176183
Rose JKC, Hadfield KA, Labavitch JM, Bennett AB (1998) Temporal sequence of cell wall dis-
assembly in rapidly ripening melon fruit. Plant Physiol 117:345361
Saab IN, Sachs MM (1996) A folding-induced xyloglucan endo-transglycosylase homolog
in maize is responsive to ethylene and associated with aerenchyma. J Plant Physiol
112:385391
Schussler EE, Longstreth DJ (2000) Changes in cell structure during the formation of root aeren-
chyma in Sagittaria lancifolia (Alismataceae). Amer J Bot 87:1219
Serpa VI, Polikarpov I (2011) Enzymes in bioenergy. In: Buckeridge MS, Goldman GH (eds)
Routes to cellulosic ethanol. Springer, New York, pp 99116
Siyiannis VF, Protonotarios VE, Zechmann B, Chorianopoulou SN, Mller M, Hawkesford MJ,
Bouranis DL (2011) Comparative spatiotemporal analysis of root aerenchyma formation pro-
cesses in maize due to sulphate, nitrate or phosphate deprivation. Protoplasma 249:671686
Soccol CR, Vandenberghe LPS, Medeiros ABP, Karp SG, Buckeridge MS, Ramos LP, Pitarelo
AP, Ferreira-Leito V, Gottschalk LMF, Ferrara MA (2010) Bioethanol from lignocelluloses:
status and perspectives in Brazil. Biores Technol 101:48204825
Steffens B, Geske T, Sauter M (2011) Aerenchyma formation in the rice stem and its promotion
by H2O2. New Phytol 190:369378
Taylor LE II, Dai Z, Decker SR, Brunecky R, Adney WS, Ding SY, Himmel ME (2008)
Heterologous expression of glycosyl hydrolases in planta: a new departure for biofuels.
Trends Biotechnol 26:413424
Tin MAS, Silva CO, Lima DU, Carpita NC, Buckeridge MS (2006) Fine structure of a
mixed-oligomer storage xyloglucan from seeds of Hymenaea courbaril. Carbohydr Polym
66:444454
Tomassini A, Sella L, Raiola A, DOvidio R, Favaronl F (2009) Characterization and expres-
sion of Fusarium graminearum endo-polygalacturonases in vitro and during wheat infection.
Plant Pathol 58:556564
Vanholme B, Cesarino I, Goeminne G, Kim H, Marroni F, Van Acker R, Vanholme R, Morreel K,
Ivens B, Pinosio S, Morgante M, Ralph J, Bastien C, Boerjan W (2013) Breeding with rare
defective alleles (BRDA): a natural Populus nigra HCT mutant with modified lignin as a case
study. New Phytol 198:765776
Verma D, Kanagaraj A, Jin S, Singh ND, Kolattukudy PE, Daniell H (2010) Chloroplast-derived
enzyme cocktails hydrolyse lignocellulosic biomass and release fermentable sugars. Plant
Biotechnol J 8:332350
Visser EJ, Bgemann GM (2006) Aerenchyma formation in the wetland plant Juncus effusus is
independent of ethylene. New Phytol 171:305314
Walton JD (1994) Deconstructing the cell wall. Plant Physiol 104:11131118
Ward R (2011) Cellulase engineering for biomass saccharification. In: Buckeridge MS, Goldman
GH (eds) Routes to cellulosic ethanol. Springer, New York, pp 137153
Willats WGT, McCartney L, Mackie W, Knox P (2001) Pectin: cell biology and prospects for
functional analysis. Plant Mol Biol 47:927
Xin Z, Watnabe N, Lam E (2011) Improving efficiency of cellulosic fermentation via genetic
engineering to create smart plants for biofuel production. Enzymes in bioenergy. In:
Buckeridge MS, Goldman GH (eds) Routes to cellulosic ethanol. Springer, New York, pp
183199
Chapter 14
Linking Plant Biology and Pretreatment:
Understanding the Structure and
Organization of the Plant Cell Wall
and Interactions with Cellulosic Biofuel
Production
Rebecca Garlock Ong, Shishir P. S. Chundawat, David B. Hodge,

Sai Keskar and Bruce E. Dale
Abstract In order to more economically process cellulosic feedstocks using a bio-

chemical pathway for fuel production, it is necessary to develop a detailed under-
standing of plant cell wall characteristics, pretreatment reaction chemistry, and
their complex interactions. However given the large number of thermochemical
pretreatment methods that are currently being researched and the extreme diversity
of plant cell wall structure and composition, this prospect is extremely challeng-
ing. Here we present the current state of research at the interface between plant
biology and pretreatment chemistry. The first two sections discuss the chemistry
of the secondary plant cell wall and how different pretreatment methods alter the
overall cell wall structure. The third section addresses how the characteristics of
the cell wall and pretreatment efficacy are impacted by different factors such as
R. G. Ong(*) S. P. Chundawat D. B. Hodge S. Keskar B. E. Dale

Department of Chemical Engineering and Materials Science, Michigan State University,
3815 Technology Blvd, Lansing, Lansing, MI 48910, USA
e-mail: garlock1@msu.edu
S. P. Chundawat
e-mail: chundawa@egr.msu.edu
R. G. Ong S. P. Chundawat D. B. Hodge S. Keskar B. E. Dale
DOE Great Lakes Bioenergy Research Center, Madison, WI, USA
S. P. Chundawat
Department of Biochemistry, University of WisconsinMadison, Madison, WI, USA
D. B. Hodge
Department of Biosystems and Agricultural Engineering, Michigan State
University, XXX, XXX
D. B. Hodge
Department of Civil, Environmental and Natural Resources Engineering, Lule
University of Technology, XXX, XXX
232 R. G. Ong et al.
plant maturity, classification, and plant fraction. The fourth section summarizes
current directions in the development of novel plant materials for improved bio-
chemical conversion. And the final section discusses the use of chemical pretreat-
ments as a screening and analysis tool for rapid identification of amenable plant
materials, and for expansion of the fundamental understanding of plant cell walls.
Keywords Enzymatic digestibility Lignocellulose Plant breeding and transgenesis

Plant cell wall Pretreatment chemistry Screening tools
Abbreviations
AFEX Ammonia fiber expansion

BMIMCl 1-butyl-3-methylimidazolium chloride
CBM Carbohydrate binding module
EMIMAc 1-ethyl-3-methylimidazolium acetate
EMIMCl 1-ethyl-3-methylimidazolium chloride
G Guaiacyl
GAX Glucuronoarabinoxylan
H p-hydroxyphenyl
IL Ionic liquid
S Syringyl
TAGs Triacylglycerols
14.1Introduction
Lignocellulosic materials are a promising source of biofuels because of their

abundance and availability. One potential conversion pathway is the biochemical
route, through enzymatic hydrolysis and fermentation of cell wall carbohydrates.
The difficulty is that although plant cell walls are permeable to small molecules,
such as water, carbon dioxide, sugars, and amino acids (Ivakov and Persson 2012),
while enzymes, with a diameter of around 51 (Ishizawa et al. 2007), are too large
to penetrate. Therefore to obtain access to polysaccharides embedded within the
cell wall in an industrially relevant time scale, some form of chemical or physical
pretreatment is needed to disrupt the cell wall structure. A large number of pre-
treatments are currently being researched (da Costa Sousa et al. 2009; Zhao et al.
2012), corresponding to a wide range of chemistries and modes of action. In addi-
tion there is enormous diversity of plant cell walls in terms of their structure and
organization (Cosgrove 2005). The chemical and physical interactions between
variables related to the feedstock (Fig.14.1a), and pretreatment (Fig.14.1b),
determines the specific types and magnitudes of effects on cell wall structure
(Fig.14.1c), and ultimately the extent of enzymatic deconstruction.
14 Linking Plant Biology and Pretreatment 233
Fig.14.1Feedstock variables (a), pretreatment variables (b), and resulting modes of action
(c)for improved enzymatic degradation of plant cell wall carbohydrates. Gray arrows in part C
represent the potential for a direct impact of one mode of action on another
14.2Secondary Cell Wall Chemistry
Higher plants have two main cell wall types, with different functions and composi-
tions. Primary walls are laid down during cell growth and elongation, and second-
ary walls are laid down after cessation of cell growth (Cosgrove 2005; Ivakov and
Persson 2012). The middle lamella is located between adjacent cells and binds them
together (Cosgrove 2005). After growth stops, lignin deposition begins in the middle
lamella and cell corners and progresses to the primary and secondary walls (Ralph et
al. 2007; Ivakov and Persson 2012). In woody plants, the primary wall is degraded
before secondary wall deposition (Jarvis 2012), however for herbaceous plants the
secondary wall is deposited directly inside the primary wall (Wilson and Hatfield
1997; Engels and Jung 1998). Not all types of cells have secondary cell walls, mainly
those requiring greater strength or rigidity (Cosgrove 2005), and some secondary
walls never lignify (Engels and Jung 1998). But because of their greater thickness,
secondary walls make up the bulk of lignocellulosic biomass and cell volume, espe-
cially in woody materials (Wilson and Hatfield 1997; Ivakov and Persson 2012).
Cellulose, hemicelluloses, and lignin are the major components of the secondary cell
wall. Cellulose forms the scaffolding of the cell wall and comprises -(14)-linked
glucan chains arranged in crystalline microfibrils. The hemicelluloses are a diverse
class of amorphous carbohydrates that cross-link cellulose microfibrils and lignin
within the cell wall. All hemicelluloses have -(14)-linked backbones of glucose
(glucans), mannose (mannans), glucose and mannose (glucomannans), or xylose
(xylans), and may be substituted with sugars, uronic acids, and acetyl groups. Lignin
is an amorphous phenylpropanoid polymer that fills in most of the remaining space
and is comprised of three different subunits that are differentiated by the number of
methoxyl groups on the phenyl ring: syringyls (S) have 2; guaiacyls (G) have 1; and
p-hydroxyphenyls (H) have 0. Pectins are another class of carbohydrate and repre-
sent a major portion of the dicot and gymnosperm primary wall, however they are
comparatively easy to extract from the cell wall or degrade (Willfr et al. 2005a, b;
DeMartini et al. 2011a). For more detailed explanations on cell wall composition and
structure of the polymers please refer to a number of reviews (Carpita and Gibeaut
1993; Ralph et al. 2007; Scheller and Ulvskov 2010; Ivakov and Persson 2012).
14.2.1Variation in Chemistry Due to Classification,

Cell Type, and Location
Bioenergy plants are grouped in three classes based on their cell wall composition:
grass-like (commelinid monocots), dicot-like (non-commelinid monocots, herba-
ceous dicots, and hardwoods), and gymnosperm (softwoods). Grass-like secondary
cell walls contain glucuronoarabinoxylan (GAX) as the main hemicellulose substi-
tuted with arabinose and some glucuronic acid, and lignin comprised of S and G
subunits with low levels of H subunits; dicot-like secondary cell walls predomi-
nantly contain glucuronoxylan substituted with 4-O-methyl-glucuronic acid and
infrequently with arabinose, and lignin comprised of similar levels of S and G subu-
nits and trace H subunits; and gymnosperm secondary walls contain slightly more
galactoglucomannan than glucuronoarabinoxylan, and lignin comprised mostly of
G subunits and low levels of H subunits (Ralph et al. 2007; Scheller and Ulvskov
2010). The type and distribution of the polymers varies within the cell and between
cell types. For all plant classifications, the cell corners and middle lamella generally
have the highest lignin content compared to the primary and secondary wall (Singh
et al. 2009; Siqueira et al. 2011; Sun et al. 2011). For corn stover, cell types can
be arranged in order of decreasing lignin and cellulose content: sclerenchyma and
tracheids>epidermis>bundle sheath>parenchyma (Sun et al. 2011). Sugarcane
follows a similar trend with lignin concentrated in the vessels followed by fiber and
parenchyma cells (Siqueira et al. 2011). Lignin in herbaceous dicots is concentrated
in the vascular ring (Wilson and Hatfield 1997; Engels and Jung 1998), but pith
parenchyma cells, though thin, are also lignified (Engels and Jung 1998).
14.2.2Covalent Linkages
The S/G ratio determines the types of inter-unit cross-linking that occur within the
lignin matrix. The -O-4 (-aryl ether) linkage is the most frequent linkage and
one of the most easily cleaved chemically (Ralph et al. 2007) and is more com-
mon in lignin containing more S subunits (Kishimoto et al. 2009). 4-O-5 linkages
are more common with a 1:1 S/G ratio, and the - linkage is more common with
a greater proportion of S subunits (Kishimoto et al. 2009; Rencoret et al. 2011).
Lignins with a greater proportion of G subunits tend to be more branched, and also
contain more chemically and thermally resistant structures (-5 and 5-5) (Ralph et
al. 2007; Kishimoto et al. 2009; Rencoret et al. 2011). As a result, hardwood lignin
is easier to degrade and has a lower glass transition temperature compared to soft-
wood lignin, which contains no syringyl subunits (Lundquist and Lundgren 1972;
Lundquist 1973; Awal and Sain 2011).
Lignin is also covalently linked to hydroxycinnamic acids, with p-coumaric
acids forming ester-linked terminal residues. Ferulic acids, which are also able to
form oligomers, are ether-linked to lignin and ester-linked to carbohydrates, either
pectins in certain dicots, or GAX arabinose side-chains in grass and dicot second-
ary walls (Iiyama et al. 1990; Harris and Trethewey 2010), though the frequency
is lower for dicots due to significantly lower arabinose substitution (Scheller and
Ulvskov 2010; Chiniquy et al. 2012). Ferulate cross-links limit enzymatic degra-
dation (Grabber et al. 1998), but the ester link with hemicellulose is easily cleaved
by most pretreatments. In addition to covalent cross-linking through hydroxycin-
namic bridges, a variety of direct cross-links have also been proposed between
lignin subunits and cell wall carbohydrates (Imamura et al. 1994; Karlsson et al.
2004; Lawoko et al. 2006).
14.2.3Non-Covalent Interactions
A great deal of interaction between cell wall polymers is in the form of hydrogen
bonding and van der Waals forces. In higher plants the glucan chains in the cel-
lulose microfibril are present predominantly in the I crystal conformation (Atalla
and Vanderhart 1984; Stone 2005). The microfibrils may interact with each other
and other cell wall polysaccharides through non-covalent interactions (Altaner
and Jarvis 2008; Ivakov and Persson 2012) and through these form aggregate-
or bundle-like structures (Donaldson 2007; Abe and Yano 2009). Glucomannans
bind more strongly to cellulose and are more resistant to extraction compared
to glucuronoxylans (Clayton and Phelps 1965; kerholm and Salmn 2001,
2004; Zhang et al. 2011a). Strength of hemicellulose binding is likely related to
interactions between the specific sugars in the hemicellulose backbone and cellu-
lose, and a recent modeling study showed fewer hydrogen bonds but greater bond
strength between cellulose and glucomannan compared to between cellulose and
xylan (Zhang et al. 2011a). However stronger binding of glucomannan may also
be related to lower side-chain substitution compared to xylan (Clayton and Phelps
1965). For the same class of hemicellulose, those with lower substitution bind
more strongly to cellulose (Whitney et al. 1998; Kabel et al. 2007; Dammstrm et
al. 2009), and the pattern of substitution also appears to have an impact (de Lima
and Buckeridge 2001). In addition to the sugar side-chains, most mannans and
xylans are acetylated (Scheller and Ulvskov 2010), which likely reduces binding
affinity towards cellulose (Altaner and Jarvis 2008). It has also been hypothesized
that hemicelluloses may be covalently linked to or embedded within cellulose
microfibrils (Cosgrove 2005).
14.3Pretreatment Chemistry
Thermochemical pretreatments alter the cell wall through chemical reactions that
cleave covalent bonds and/or disrupt non-covalent interactions between cell wall
polymers (Fig.14.2) as well as through thermal softening and solubilization of
biomass components. These chemical changes in combination with the physi-
cal removal and/or relocalization of cell wall components cause structural changes
that improve enzymatic digestibility. Most pretreatments can be grouped based on
their general effect on cell wall structure: those that remove lignin (alkaline/oxida-
tive), those that remove hemicellulose and relocalize lignin (acidic), and those that
fractionate cell wall components (ionic liquid, organosolv, and phosphoric acid)
(Fig. 14.2). Most pretreatments, except for biological pretreatments and ionic
liquids (ILs), can also be arranged in a continuum based on the nucleophilicity/electro-
philicity of their main reactants (Fig.14.3). Though less precise, the continuum can
also be thought of in terms of pH (Pedersen and Meyer 2010; Garlock et al. 2011).
Almost all of these pretreatments cleave some fraction of acetyl groups from the
hemicellulose backbone (Maloney et al. 1985; Kumar et al. 2009; Shi et al. 2011)
and use conditions that break -ether linkages in lignin (Saake and Lehnen 2007).
The main mode of action for alkaline and oxidative pretreatments is through
nucleophilic substitution and/or oxidation of esters and -ethers within lignin and
between cell wall polymers (Tarkow and Feist 1969; Iiyama et al. 1990; Sewalt
et al. 1996). At very high alkali concentrations, carbohydrate monomers can be
removed via peeling reactions and converted to acids (e.g. lactic acid) (Knill
and Kennedy 2003). As reactant concentration and temperature decrease, peel-
ing reactions become less likely to occur and fewer -aryl-ether bonds are bro-
ken. Ammonia, a weaker nucleophile, does not cleave -ethers but is known to
cleave ester-linkages between hemicellulose and hydroxycinnamic acids (Wang
et al. 1964; Azarpira et al. 2011). In contrast, acidic pretreatments mainly act
through electrophilic hydrolysis of ester cross-links, -ether bonds, and glycosidic
Fig.14.2Main molecular scale (chemical) impacts to plant cell wall components by ther-
mochemical pretreatments and, in conjunction with mass transfer of biomass components, the
resulting nanoscale (structural) changes for the three main classes of pretreatment
linkages, and they can also catalyze the dehydration of monomeric sugars. Room
temperature acid treatment is able to break ether linkages between hydroxycin-
namic acids and lignin/hemicelluloses (Wallace et al. 1995); while higher tem-
peratures are needed to hydrolyze esters (Sannigrahi et al. 2009). Though -ether
bonds can be broken by strong acidic pretreatments, they are more readily hydro-
lyzed by alkali (Saake and Lehnen 2007). The key feature of acidic pretreatments
is the hydrolysis of glycosyl linkages that allows for extraction of hemicellulose-
derived oligomers and monomers. Xylans are more easily hydrolyzed than man-
nans (McGee and April 1982; Tunc and van Heiningen 2008; Vrnai et al. 2010),
and for side-chains, arabinose is more easily removed than galactose, and galac-
tose than 4-O-methyl-glucuronic acid (McGee and April 1982; Sun and Cheng
2005). During hydrothermal pretreatments, hydronium ion concentration is
Fig.14.3Thermochemical pretreatments arranged in order of reactant nucleophilicity and effect

on cell wall covalent linkages
initially governed by water autoionization and later by release of weak, biomass-

derived acids (Garrote et al. 1999), and for these weakly acidic pretreatments,
most hemicellulose is released as oligomers, only the most amenable side-chains
are cleaved, and most lignin inter-unit linkages remain intact (Garlock et al. 2011).
A number of pretreatments (AFEX1, liquid hot water, dilute acid, and acid-cata-
lyzed organosolv) have also been shown to deposit lignin-rich globules on the surface
of the cell wall (Donohoe et al. 2008; Chundawat et al. 2011; Donohoe etal. 2011;
Koo et al. 2012). For acidic pretreatments, particularly those catalyzed by sulfuric
acid, lignin can condense and form new bonds (Xiao et al. 2013; Lundquist 1973;
Karlsson et al. 1988). Most pretreatments also generate degradation compounds that
influence downstream processes, and the specific compounds that are formed are
determined by the interaction of plant cell wall chemistry (grass, hardwood, or soft-
wood) with pretreatment chemistry (Chundawat et al. 2010; Du et al. 2010).
In addition to cleavage of covalent bonds, some pretreatments (sodium hydrox-
ide, liquid ammonia, phosphoric acid, and ILs) disrupt hydrogen bonding within
cellulose microfibrils and generate more digestible forms of cellulose (amor-
phous>II, III>I). The main mode of action for ionic liquids is the disruption of
hydrogen bonding and decrystallization of cellulose to the extent that fractionation
and lignin removal may not be necessary for high enzymatic conversions (Wu etal.
2011). There are indications that IL reactivity is related to both the ability of the
1 AFEX is a registered trademark of MBI International, Lansing, MI.

anion to accept hydrogen bonds (Tadesse and Luque 2011; Gericke et al. 2012; King
et al. 2012) and the length of the alkyl substituent chain on the cation, with shorter
chain lengths leading to more effective cellulose dissolution (Zhang et al. 2005).
Combined, the anion and cation form an electron donoracceptor matrix with the
cellulose hydroxyl groups that facilitates dissolution (Tadesse and Luque 2011; Xu
et al. 2012). In ILs, pH is a measure of dissociation between the anion and cation
(MacFarlane et al. 2006) and anions and cations can be classified as acidic, basic, or
neutral. For example, the imidazolium ring has acidic properties that are believed to
result in acid catalytic effects (MacFarlane et al. 2006). An IL with an acidic cation
and basic anion like 1-ethyl-3-methylimidazolium acetate (EMIMAc) has a larger
degree of dissociation (~pH 11) (Singh et al. 2009; Muhammad et al. 2012), which
is likely related to its ability to both decrystallize cellulose and dissolve lignin.
Imidiazolium ILs with a weakly basic anion like 1-butyl-3-methylimidizolium
chloride (BMIMCl) and 1-ethyl-3-methylimidizolium chloride (EMIMCl) (~pH 6)
are more selective for dissolving cellulose (Zhang et al. 2013b). The IL anion may
also enhance catalytic reactions, and ILs with anions that are less basic than water
(e.g. Cl) turn strong acids into weaker acids, however ILs with anions that are
more basic than water (e.g. acetate) turn weak acids (like water and acetic acid) into
stronger acids (MacFarlane et al. 2006). This may be one reason for the beneficial
effect of water observed in EMIMAc, though this may also be related to reductions
in viscosity (Fu and Mazza 2011). IL viscosity, which is much higher than conven-
tional solvents, impacts cellulose dissolution through mass transfer and this can be
difficult to separate from kinetic impacts (Gericke et al. 2012). The effectiveness
of an IL is also dependent on temperature. Pure cellulose dissolves in imidazolium
ILs between 80 and 100C (Zhang et al. 2005; Vitz et al. 2009), but pretreatment
of whole biomass requires higher temperatures (~130C) for significant decrystal-
lization of undissolved fractions (Kimon et al. 2011), which might be related to the
glass transition temperature of lignin (Keskar et al. 2011; Li et al. 2011).
14.4Impacts of Plant Characteristics on Cell Wall

Degradation
14.4.1Plant Classification
Different pretreatments process certain classifications of plants more effectively

than others (Wyman et al. 2013), however, for the same pretreatment method, plant
materials can almost always be arranged in the following order, either with regard
to digestibility for the same conditions, or severity of conditions required for equiv-
alent digestibility: grasses>herbaceous dicots>hardwoods>softwoods (Arantes
and Saddler 2011; DeMartini and Wyman 2011a; Garlock et al. 2012b). This order
is largely due to four factors that increasingly hinder pretreatment reaction kinetics
and mass transfer: (1) increase in proportion of recalcitrant covalent linkages within
the cell wall (esters ->ethers ->carboncarbon bonds); (2) increase in strength of
hydrogen-bonding of major hemicellulose sugars to cellulose; (3) increase in average

cell wall thickness and proportion of the cell volume occupied by cell wall; and
(4) increase in the proportion of lignin versus cellulose, although if cellulose accessibil-
ity is sufficiently increased, the actual presence of lignin during hydrolysis is not a major
issue (Jeoh et al. 2007; Chundawat et al. 2011; Rollin et al. 2011; Wiman et al. 2012).
14.4.2Plant Varieties
A handful of studies have looked at differences in digestibility and yields for cul-
tivars within the same species. Upland and lowland switchgrass when harvested
around the same time in the same location had similar sugar yields for most pre-
treatment methods (Kim et al. 2011) and similar optimal pretreatment conditions
and enzyme loading (Garlock et al. 2012a). Results for wheat straw were varied,
with one study that indicated sugar yields (g/g dry biomass) from hydrothermally
pretreated wheat straw were not influenced by cultivar (Larsen et al. 2012), while
two other studies found a significant variation in sugar yields across all cultivars
(Lindedam et al. 2010; Lindedam et al. 2012).
14.4.3Plant Cell Types and Tissues
Herbaceous feedstocks can show major differences in conversion between dif-

ferent portions of the plant or different cell types that may influence practical
considerations such as harvesting methods and fractionation prior to pretreat-
ment. In general, pith tends to be more digestible than the vascular bundles and
the rind/epidermis. One study found that sugar yields follow the same pat-
tern of digestibility for both hydrothermally pretreated and untreated materi-
als (pith>leaves>rind) (Zeng et al. 2012). Pith cells of sugar cane were highly
digestible by enzymes even without pretreatment and following chlorite treatment
the rind cells became significantly more digestible (Siqueira et al. 2011).
For herbaceous botanical fractions, the general trend is that stems are easier to
digest than leaves. For AFEX-pretreatment, corn fractions were more digestible
in order of decreasing lignin content (husk>leaf>stem>cob) (Garlock etal.
2009). For sodium hydroxide pretreatment, corn stover fractions released the most
glucan in order of husks, cob, and leaves>upper stem>lower stem (Duguid et
al. 2009) and corn stover and wheat straw fractions that contained more lignin
showed a greater improvement with a higher catalyst loading (Duguid et al. 2007,
2009). Hydrothermally pretreated grasses and legume stems had lower percent
sugar conversions than leaves, but higher total sugars released (DeMartini and
Wyman 2011a). Miscanthus fractions showed decreasing cellulose conversion
with: leaves>sheath>stem (Le Ngoc Huyen et al. 2010).
14.4.4Harvest Date and Maturity
For herbaceous crops that have annual growth cycles, harvest date significantly
impacts composition and biomass yields. As the plant approaches full maturity and
senescence, the relative proportion of lignin and structural carbohydrates increase
with a simultaneous decrease in soluble sugars, protein, and minerals (Dien et al.
2006). Harvest during the growing season can result in a highly digestible mate-
rial, but one that also has significant nitrogen and ash content (Bals et al. 2010),
which can impact farm economics, sustainability, and conversions. Some studies
have shown little impact on total sugars released due to maturity (Dien et al. 2006;
Garlock et al. 2009). However, there is a consistent decrease in digestibility and
biomass yields when harvest is delayed from fall to winter or spring, largely due
to loss of leaves and other fragile, digestible portions of the plant (Pordesimo et al.
2005; Adler et al. 2006; Le Ngoc Huyen et al. 2010; Kim et al. 2011). With regard
to woody materials, one paper examined sugar yields from different growth rings
and found no significant variation between mature wood and juvenile wood, despite
an increase in lignin content with age of the ring (DeMartini and Wyman 2011b).
14.4.5Composition
The most common trend reported for the effect of biomass composition on hydrol-
ysis yields, is that glucan digestibility is negatively correlated to total lignin con-
tent (Davison et al. 2006; Dien et al. 2006; Rock et al. 2009; Garlock et al. 2012b).
Lignin monomer composition may also be important, as a decrease in the S/G
ratio leads to more recalcitrant linkages, and pretreatments that can break them
would be expected to show a higher digestibility compared to those that do not.
However, based on a number of studies S/G ratio may or may not be correlated to
improved digestibility, depending on other plant cell wall properties and whether
and how the plant was pretreated (Chen et al. 2002; Mechin et al. 2005; Davison
etal. 2006; Li et al. 2010; Studer et al. 2011b; Zhang et al. 2011b).
14.5Designing Improved Feedstocks
A number of strategies for developing plants designed for deconstruction have

been reviewed in recent years (Carpita 2012; Jung et al. 2012; Abramson et al.
2013). These strategies can be grouped broadly as altering lignin (content, monol-
ignol composition, and degree of polymerization), increasing and/or altering poly-
saccharides (content, composition, or crystallinity), expressing cell wall-degrading
or modifying enzymes in planta, or producing oils in vegetative tissues.
14.5.1Alterations to Lignin
Initial studies on plants with altered lignin contents began with the brown mid-
rib mutations (Barrire et al. 2004) for improved ruminant digestibility. Plant
lines have subsequently been engineered with decreased and altered lignin con-
tent by changing the expression of monolignol biosynthetic enzymes. Decreasing
expression of one or more of the monolignol synthesis enzymes has been shown
to decrease total lignin content and improve the enzymatic digestibility of alfalfa
following hot water pretreatment (Chen and Dixon 2007). However, decreasing the
total lignin content of the cell wall also impairs the overall fitness of the plant and
can lead to dwarfed plants and failure to accumulate biomass (Casler et al. 2002;
Voelker et al. 2011). As a consequence of this, more recent strategies have been
focused on altering the ratio of monolignols, and increasing the S/G ratio in hybrid
poplar has been shown to improve alkaline delignification (Stewart et al. 2009) and
digestibility following alkaline and dilute acid pretreatment, though there was no
significant difference following AFEX treatment (Ong 2011). Increasing S/G in
Arabidopsis was shown to improve the enzymatic release of glucose following hot
water pretreatment (Li et al. 2010). A recent study found that decreasing total lignin
content concurrently with decreasing S/G in switchgrass improved the enzymatic
glucose yield following dilute acid pretreatment, as well as decreasing pretreatment
severity and cellulase loadings, and increasing ethanol yield (Fu et al. 2011).
Another strategy has been to introduce novel monolignols or proteins that make
the cell wall more amenable to chemical deconstruction without impacting total
lignin content or plant fitness. These approaches, all of which have been shown to
increase digestibility and/or lignin removal to some extent include adding monol-
ignols that shorten the degree of polymerization (p- hydroxybenzyaldehydes) (Eudes
et al. 2012), monolignols that incorporate alkali-labile ester linkages within the lignin
matrix, e.g. novel ester-based di-lignols as lignin precursors (Grabber et al. 2008;
Simmons et al. 2010), and glycoproteins that participate in cross-couplings with
lignin, such as tyrosine-rich hydroxyproline-rich glycoprotein (Liang et al. 2008).
14.5.2Alterations to Polysaccharides
Altering cell wall polysaccharides is another method to reduce cell wall recalci-
trance or increase the amount of substrate. One strategy is to decrease cellulose
crystallinity by overexpressing cellulose synthases with impaired functional-
ity (Harris et al. 2009) or by overexpressing a membrane-bound endoglucanase,
KORRIGAN (Maloney and Mansfield 2010). Another strategy is to increase the
carbohydrate content of the plant cell wall. Cellulose content and crystallinity
increased in poplar by over-expressing a sucrose synthase gene (Coleman et al.
2006) and various amorphous polysaccharides have been targeted for accumula-
tion, including starch (Chuck et al. 2011 and mixed-linkage -glucans (Pauly et
al. 2011). In contrast, reductions in glucuronoxylan content in poplar showed an
increase in digestibility by enzymes alone (Lee et al. 2009). In rice, loss of activity
for a xylosyltransferase thought be responsible for arabinosyl substitution of the
xylan backbone resulted in a slight increase in arabinose substitution and decrease
in hydroxycinnamic acid content, resulting in increased extractability of xylan and
enzymatic digestibility (Chiniquy et al. 2012). Alteration of O-acetylation of hemi-
celluloses may also lead to a decrease in acetate content for reduced inhibition of
fermentation or altered capacity of hemicelluloses to hydrogen bond with other
cell wall polymers (Gille and Pauly 2012). Other work has demonstrated improved
enzymatic digestibility of plant cell walls by preventing de-methyl esterification in
the pectin homogalacturonan (Lionetti et al. 2010), which limits the ability to form
Ca2+-mediated cross-links, increasing primary cell wall porosity and decreasing
rigidity and cell-to-cell adhesion in primary cell walls.
14.5.3Transcription Factors for Secondary Cell Wall

Formation
Regulatory networks have recently been identified comprising several transcrip-

tion factors that act as master switches responsible for controlling the temporal
and spatial regulation of collections of genes involved in the secondary cell wall
synthesis, assembly, and thickening (Shen et al. 2012). One study ectopically over-
expressed a MYB transcription factor in switchgrass to down-regulate the genes
associated with monolignol biosynthetic pathways and identified phenotypic out-
comes of reduced lignin and reduced p-coumarate to ferulate ratios that resulted
in a tripling of enzymatic sugar release (Shen et al. 2012). Other work identified a
mutation in WRKY transcription factors to be responsible for secondary cell wall
thickening and significantly increased cellulose, hemicellulose, and lignin depo-
sition in the pith cells of model dicots, increasing the overall plant density, and
potentially providing a route for increasing accumulation of fermentable sugars in
plant cell walls (Verma et al. 2010).
14.5.4Expression of Cell Wall Degrading Enzymes in Planta
The high cost and doses of enzymes required for cellulosic biofuels are critical eco-
nomic barriers for commercialization. Expression of thermophilic cellulases in
the apoplast (Sticklen 2006) or mesophilic cellulases in chloroplasts (Verma et al.
2010) are one possible route for generating some of the cellulolytic enzymes in situ.
Cellulolytic enzymes can be generated in planta to supplement other enzymes during
hydrolysis, however even mild pretreatment of the biomass can significantly lower
their activity (Teymouri et al. 2004). Expression of feruloyl esterases in grasses which
cleave ferulate ester cross-links has been found to improve both enzymatic and in
vitro ruminant digestibilities (Buanafina et al. 2008). Expression of plant cellulolytic
enzymes that are active under plant physiological conditions (Hartati et al. 2008) or
cellulose binding modules (CBMs) (Shoseyov et al. 2006) in the apoplast have been
found to increase growth and biomass accumulation, presumably by increased cell
wall loosening, but with the potential disadvantage of impaired plant fitness.
14.5.5Production of Oils in Vegetative Tissues
One way to increase the energy content of lignocellulosic biomass is to modify

plants to produce oils, fatty acids, or triacylglycerols (TAGs) in vegetative tissues
(Durrett et al. 2008). In one study triacylglycerols were accumulated in senesc-
ing Arabidopsis leaves by either blocking fatty acid breakdown, or by ectopically
expressing the LEC2 seed development transcription factor in leaves (Slocombe
etal. 2009). Another study successfully shifted the carbon flux in Arabidopsis
leaves from starch biosynthesis to the production and accumulation of triacyl-
glycerols by simultaneously reducing the expression of a catalytic subunit of
ADP-glucose pyrophosphorylase and ectopically expressing the WRINKLED1
transcription factor that is involved in seed oil biosynthesis (Sanjaya et al. 2011).
14.6Pretreatment as a Screening and Analysis Tool:

Expanding Our Understanding of the Plant Cell Wall
Re-engineering plants to provide phenotypic traits desirable of an ideal biofuel

energy crop is an area of intense research, as highlighted previously. However, it
is vitally important to evaluate processing capabilities of new materials as they are
being generated, as biomass recalcitrance may not favorably correlate with the traits
selected for during transgenesis or breeding. With recent advances in high-through-
put analytical techniques, it is now feasible to quickly screen for desirable traits from
very large libraries of biomass phenotypes, while requiring only small sample quanti-
ties for detailed analyses. In addition to screening, high-throughput techniques are
also helping to further understanding of the relationship between biomass conver-
sion and plant cell wall characteristics. For example, high throughput composition
analysis techniques allowed for screening of thousands of poplar samples for lignin
content and S/G ratios, and from this a fairly large subset was further tested using a
high-throughput pretreatment and enzymatic hydrolysis method in order to determine
the relative impacts of lignin and S/G ratio on sugar yields (Studer et al. 2011b).
As in the example above, high-throughput pretreatments can now be carried out in
custom-designed microplate-based reactors that have been developed for both acidic
and alkaline pretreatments (Santoro et al. 2010; Selig et al. 2010; Studer et al. 2011a).
Rapid, small-scale compositional analysis methods are able to determine cell wall
composition, both before and after pretreatment (DeMartini et al. 2011b; Selig et al.
2011). These techniques can be coupled to medium/high-throughput analyses using
LCMS/MS and 2D-NMR for more detailed elucidation of changes in cell wall struc-
ture, composition, and degradation (Chundawat et al. 2008; Kim and Ralph 2010;
Morreel et al. 2010). Semi-automated (medium/low throughput) electron microg-

raphy and immunolabeling based techniques have also been used in recent years to
characterize the complex interplay of pretreatment severity and cell wall ultra-struc-
tural modifications (Donohoe et al. 2009; Pattathil et al. 2010; Chundawat et al. 2011;
Zhang et al. 2013a). To this end a bio-analytic toolkit was developed, comprising
more than 200 glycan-directed monoclonal antibodies that recognize distinct epitopes
present on various categories of plant cell wall polysaccharides (Pattathil et al. 2010).
This microplate-based, quantitative assay has provided insights into the relationship
between pretreatment severity and cell wall polysaccharide accessibility and extrac-
tion, and the molecular architecture of the plant cell wall (Alonso-Simn et al. 2010;
DeMartini et al. 2011a). As indicated by Moller et al. (2007), monoclonal antibod-
ies directed against cell wall glycans provides complementary compositional data that
could be used to optimize pretreatment conditions and enzyme cocktails necessary for
more efficient degradation of lignocellulose.
The effectiveness of pretreatments on bioconversion has been evaluated using
micro-scale based rapid enzymatic hydrolysis (Chundawat et al. 2008; Banerjee
et al. 2010; Gomez et al. 2010; Jger et al. 2011; Riedlberger and Weuster-Botz
2012) and microbial fermentation based assays (Funke et al. 2010; Riedlberger and
Weuster-Botz 2012). These assays can be coupled with microplate-based pretreat-
ments to facilitate rapid screening of several hundred biomass specimens (Studer et
al. 2010). Additionally, with developments in micro-scale cell-free protein expres-
sion systems it is possible to selectively optimize enzyme combinations necessary
for different pretreatments and biomass types (Chandrasekaran et al. 2010).
14.7Conclusions
In recent years understanding of the chemistry and structure of the plant cell wall
has progressed rapidly. Pretreatment research has contributed to understanding of the
distribution and composition of various cell wall polysaccharides within the many
different classes of cell walls. Future work will continue to delve more deeply into
the complex relationships between cell wall and pretreatment chemistry to improve
and develop novel conversion methods for release of cell wall sugars and to improve
biomass characteristics for conversion to biofuels. High-throughput analytical tech-
niques and tools that allow for rapid analysis of small quantities of samples will
allow for more efficient comparisons in the development of new feedstocks and
processing methods, and improved understanding of the fundamental relationships
between cell wall chemistry and structure and pretreatment chemistry.
References
Abe K, Yano H (2009) Comparison of the characteristics of cellulose microfibril aggregates of

wood, rice straw and potato tuber. Cellulose 16:10171023
Abramson M, Shoseyov O, Hirsch S, Shani Z (2013) Genetic modifications of plant cell walls to
increase biomass and bioethanol production. In: Lee JW (ed) Advanced biofuels and bio-
products. Springer, New York, pp 315338
Adler PR, Sanderson MA, Boateng AA, Weimer PJ, Jung H-JG (2006) Biomass yield and biofuel
quality of switchgrass harvested in fall or spring. Agron J 98:15181525
kerholm M, Salmn L (2001) Interactions between wood polymers studied by dynamic ft-ir
spectroscopy. Polym 42:963969
kerholm M, Salmn L (2004) Softening of wood polymers induced by moisture studied by
dynamic FTIR spectroscopy. J Appl Polym Sci 94:20322040
Alonso-Simn A, Kristensen JB, bro J, Felby C, Willats WGT, Jrgensen H (2010) High-
throughput microarray profiling of cell wall polymers during hydrothermal pre-treatment of
wheat straw. Biotechnol Bioeng 105:509514
Altaner CM, Jarvis MC (2008) Modelling polymer interactions of the `molecular velcro type in
wood under mechanical stress. J Theor Biol 253:434445
Arantes V, Saddler J (2011) Cellulose accessibility limits the effectiveness of minimum cellu-
lase loading on the efficient hydrolysis of pretreated lignocellulosic substrates. Biotechnol
Biofuel 4(3):116
Atalla RH, Vanderhart DL (1984) Native cellulose: a composite of two distinct crystalline forms.
Science 223:283285
Awal A, Sain M (2011) Spectroscopic studies and evaluation of thermorheological properties of
softwood and hardwood lignin. J Appl Polym Sci 122:956963
Azarpira A, Lu F, Ralph J (2011) Reactions of dehydrodiferulates with ammonia. Org Biomol
Chem 9:67796787
Bals B, Rogers C, Jin MJ, Balan V, Dale B (2010) Evaluation of ammonia fibre expansion
(AFEX) pretreatment for enzymatic hydrolysis of switchgrass harvested in different seasons
and locations. Biotechnol Biofuel 3(1):111
Banerjee G, Car S, Scott-Craig J, Borrusch M, Walton J (2010) Rapid optimization of enzyme
mixtures for deconstruction of diverse pretreatment/biomass feedstock combinations.
Biotechnol Biofuel 3:22
Barrire Y, Ralph J, Mchin V, Guillaumie S, Grabber JH, Argillier O et al (2004) Genetic and
molecular basis of grass cell wall biosynthesis and degradability. II. Lessons from brown-
midrib mutants. C R Biol 327:847860
Buanafina MMdO, Langdon T, Hauck B, Dalton S, Morris P (2008) Expression of a fungal feru-
lic acid esterase increases cell wall digestibility of tall fescue (Festuca arundinacea). Plant
Carpita NC (2012) Progress in the biological synthesis of the plant cell wall: New ideas for
improving biomass for bioenergy. Curr Opin Biotechnol 23:330337
Carpita NC, Gibeaut DM (1993) Structural models of primary cell walls in flowering plants:
Consistency of molecular structure with the physical properties of the walls during growth.
Plant J 3:130
Casler MD, Buxton DR, Vogel KP (2002) Genetic modification of lignin concentration affects
fitness of perennial herbaceous plants. Theor Appl Genet 104:127131
Chandrasekaran A, Bharadwaj R, Park JI, Sapra R, Adams PD, Singh AK (2010) A microscale
platform for integrated cell-free expression and activity screening of cellulases. J Proteome
Res 9:56775683
Chen F, Dixon RA (2007) Lignin modification improves fermentable sugar yields for biofuel
production. Nat Biotechnol 25:759761
Chen L, Auh C, Chen F, Cheng XF, Aljoe H, Dixon RA et al (2002) Lignin deposition and asso-
ciated changes in anatomy, enzyme activity, gene expression, and ruminal degradability in
stems of tall fescue at different developmental stages. J Agric Food Chem 50:55585565
Chiniquy D, Sharma V, Schultink A, Baidoo EE, Rautengarten C, Cheng K et al (2012)
XAX1 from glycosyltransferase family 61 mediates xylosyltransfer to rice xylan. PNAS
109:1711717122
Chuck GS, Tobias C, Sun L, Kraemer F, Li C, Dibble D, et al (2011) Overexpression

of the maize Corngrass1 microRNA prevents flowering, improves digestibility,
and increases starch content of switchgrass. PNAS, http://www.pnas.org/content/
early/2011/10/04/1113971108.abstract. (Ahead of Print)
Chundawat SPS, Balan V, Dale BE (2008) High-throughput microplate technique for enzymatic
hydrolysis of lignocellulosic biomass. Biotechnol Bioeng 99:12811294
Chundawat SPS, Donohoe BS, Sousa L, Elder T, Agarwal UP, Lu F et al (2011) Multi-scale visu-
alization and characterization of plant cell wall deconstruction during thermochemical pre-
treatment. Energ Environ Sci 4:973984
Chundawat SPS, Vismeh R, Sharma LN, Humpula JF, da Costa Sousa L, Chambliss CK et al
(2010) Multifaceted characterization of cell wall decomposition products formed during
ammonia fiber expansion (AFEX) and dilute acid based pretreatments. Bioresour Technol
101:84298438
Clayton DW, Phelps GR (1965) The sorption of glucomannan and xylan on -cellulose wood fib-
ers. J Polym Sci: Part C 11:197220
Coleman HD, Ellis DD, Gilbert M, Mansfield SD (2006) Up-regulation of sucrose synthase and
udp-glucose pyrophosphorylase impacts plant growth and metabolism. Plant Biotechnol J
4:87101
Cosgrove DJ (2005) Growth of the plant cell wall. Nat Rev Mol Cell Biol 6:850861
da Costa Sousa L, Chundawat SPS, Balan V, Dale BE (2009) `Cradle-to-grave assessment of
existing lignocellulose pretreatment technologies. Curr Opin Biotechnol 20:339347
Dammstrm S, Salmn L, Gatenholm P (2009) On the interactions between cellulose and xylan,
a biomimetic simulation of the hardwood cell wall. BioResources 4:314
Davison BH, Drescher SR, Tuskan GA, Davis MF, Nghiem NP (2006) Variation of S/G ratio and
lignin content in a Populus family influences the release of xylose by dilute acid hydrolysis.
Appl Biochem Biotechnol 129132:427435
de Lima DU, Buckeridge MS (2001) Interaction between cellulose and storage xyloglucans: the
influence of the degree of galactosylation. Carbohydr Polym 46:157163
DeMartini J, Wyman C (2011a) Composition and hydrothermal pretreatment and enzymatic sac-
charification performance of grasses and legumes from a mixed-species prairie. Biotechnol
Biofuel 4(52):110
DeMartini JD, Pattathil S, Avci U, Szekalski K, Mazumder K, Hahn MG et al (2011a)
Application of monoclonal antibodies to investigate plant cell wall deconstruction for biofu-
els production. Energ Environ Sci 4:43324339
DeMartini JD, Studer MH, Wyman CE (2011b) Small-scale and automatable high-throughput
compositional analysis of biomass. Biotechnol Bioeng 108:306312
DeMartini JD, Wyman CE (2011b) Changes in composition and sugar release across the annual
rings of Populus wood and implications on recalcitrance. Bioresour Technol 102:13521358
Dien BS, Jung H-JG, Vogel KP, Casler MD, Lamb JFS, Iten L et al (2006) Chemical composi-
tion and response to dilute-acid pretreatment and enzymatic saccharification of alfalfa, reed
canarygrass, and switchgrass. Biomass Bioenerg 30:880891
Donaldson L (2007) Cellulose microfibril aggregates and their size variation with cell wall type.
Wood Sci Technol 41:443460
Donohoe B, Decker S, Tucker M, Himmel M, Vinzant T (2008) Visualizing lignin coalescence
and migration through maize cell walls following thermochemical pretreatment. Biotechnol
Bioeng 101:913925
Donohoe BS, Selig MJ, Viamajala S, Vinzant TB, Adney WS, Himmel ME (2009) Detecting cel-
lulase penetration into corn stover cell walls by immuno-electron microscopy. Biotechnol
Bioeng 103:480489
Donohoe BS, Vinzant TB, Elander RT, Pallapolu VR, Lee YY, Garlock RJ et al (2011) Surface
and ultrastructural characterization of raw and pretreated switchgrass. Bioresour Technol
102:1109711104
Du BW, Sharma LN, Becker C, Chen S-F, Mowery RA, van Walsum GP et al (2010) Effect of
varying feedstockpretreatment chemistry combinations on the formation and accumulation
of potentially inhibitory degradation products in biomass hydrolysates. Biotechnol Bioeng

107:430440
Duguid KB, Montross MD, Radtke CW, Crofcheck CL, Shearer SA, Hoskinson RL (2007)
Screening for sugar and ethanol processing characteristics from anatomical fractions of
wheat stover. Biomass Bioenerg 31:585592
Duguid KB, Montross MD, Radtke CW, Crofcheck CL, Wendt LM, Shearer SA (2009) Effect of
anatomical fractionation on the enzymatic hydrolysis of acid and alkaline pretreated corn
stover. Bioresour Technol 100:51895195
Durrett TP, Benning C, Ohlrogge J (2008) Plant triacylglycerols as feedstocks for the production
of biofuels. Plant J 54:593607
Engels FM, Jung HG (1998) Alfalfa stem tissues: cell-wall development and lignification. Ann
Bot 82:561568
Eudes A, George A, Mukerjee P, Kim JS, Pollet B, Benke PI et al (2012) Biosynthesis and incor-
poration of side-chain-truncated lignin monomers to reduce lignin polymerization and
enhance saccharification. Plant Biotechnol J 10:609620
Fu C, Mielenz JR, Xiao X, Ge Y, Hamilton CY, Rodriguez M, et al (2011) Genetic manipulation
of lignin reduces recalcitrance and improves ethanol production from switchgrass. PNAS,
http://www.pnas.org/content/early/2011/02/04/1100310108.abstract. (Ahead of Print)
Fu D, Mazza G (2011) Aqueous ionic liquid pretreatment of straw. Bioresour Technol
102:70087011
Funke M, Buchenauer A, Mokwa W, Kluge S, Hein L, Muller C et al (2010) Bioprocess control
in microscale: scalable fermentations in disposable and user-friendly microfluidic systems.
Microb Cell Fact 9(86):113
Garlock R, Chundawat S, Balan V, Dale B (2009) Optimizing harvest of corn stover fractions
based on overall sugar yields following ammonia fiber expansion pretreatment and enzy-
matic hydrolysis. Biotechnol Biofuel 2(29):114
Garlock RJ, Balan V, Dale BE (2012a) Optimization of AFEX pretreatment conditions
and enzyme mixtures to maximize sugar release from upland and lowland switchgrass.
Garlock RJ, Balan V, Dale BE, Ramesh Pallapolu V, Lee YY, Kim Y et al (2011) Comparative
material balances around pretreatment technologies for the conversion of switchgrass to sol-
uble sugars. Bioresour Technol 102:1106311071
Garlock RJ, Bals B, Jasrotia P, Balan V, Dale BE (2012b) Influence of variable species composi-
tion on the saccharification of AFEX pretreated biomass from unmanaged fields in com-
parison to corn stover. Biomass Bioenerg 37:4959
Garrote G, Domnguez H, Paraj JC (1999) Hydrothermal processing of lignocellulosic materi-
als. Eur J Wood Wood Prod 57:191202
Gericke M, Fardim P, Heinze T (2012) Ionic liquidspromising but challenging solvents for
homogeneous derivatization of cellulose. Molecules 17:74587502
Gille S, Pauly M (2012) O-acetylation of plant cell wall polysaccharides. Front Plant Sci
3(12):17
Gomez L, Whitehead C, Barakate A, Halpin C, McQueen-Mason S (2010) Automated sac-
charification assay for determination of digestibility in plant materials. Biotechnol Biofuel
3(23):112
Grabber JH, Hatfield RD, Lu FC, Ralph J (2008) Coniferyl ferulate incorporation into
lignin enhances the alkaline delignification and enzymatic degradation of cell walls.
Biomacromolecules 9:25102516
Grabber JH, Ralph J, Hatfield RD (1998) Ferulate cross-links limit the enzymatic degradation of
synthetically lignified primary walls of maize. J Agric Food Chem 46:26092614
Harris D, Stork J, Debolt S (2009) Genetic modification in cellulose-synthase reduces crystal-
linity and improves biochemical conversion to fermentable sugar. GCB Bioenergy 1:5161
Harris P, Trethewey J (2010) The distribution of ester-linked ferulic acid in the cell walls of angi-
osperms. Phytochem Rev 9:1933
Hartati S, Sudarmonowati E, Park YW, Kaku T, Kaida R, Baba Ki et al (2008) Overexpression of

poplar cellulase accelerates growth and disturbs the closing movements of leaves in sengon.
Iiyama K, Lam TBT, Stone BA (1990) Phenolic acid bridges between polysaccharides and lignin
in wheat internodes. Phytochem 29:733737
Imamura T, Watanabe T, Kuwahara M, Koshijima T (1994) Ester linkages between lignin
and glucuronic acid in lignin-carbohydrate complexes from Fagus crenata. Phytochem
37:11651173
Ishizawa CI, Davis MF, Schell DF, Johnson DK (2007) Porosity and its effect on the digestibility
of dilute sulfuric acid pretreated corn stover. J Agric Food Chem 55:25752581
Ivakov A, Persson S (2012) Plant cell walls. eLS. Wiley, pp 117 doi: 10.1002/9780470015902.
a0001682.pub2
Jger G, Wulfhorst H, Zeithammel EU, Elinidou E, Spiess AC, Bchs J (2011) Screening of cel-
lulases for biofuel production: online monitoring of the enzymatic hydrolysis of insoluble
cellulose using high-throughput scattered light detection. Biotechnol J 6:7485
Jarvis MC (2012) Sclerenchyma. eLS. Wiley, pp 13 doi: 10.1002/9780470015902.a0002082.pub2
Jeoh T, Ishizawa CI, Davis MF, Himmel ME, Adney WS, Johnson DK (2007) Cellulase digest-
ibility of pretreated biomass is limited by cellulose accessibility. Biotechnol Bioeng
98:112122
Jung H-JG, Samac DA, Sarath G (2012) Modifying crops to increase cell wall digestibility. Plant
Sci 185186:6577
Kabel MA, van den Borne H, Vincken J-P, Voragen AGJ, Schols HA (2007) Structural differ-
ences of xylans affect their interaction with cellulose. Carbohydr Polym 69:94105
Karlsson O, Ikeda T, Kishimoto T, Magara K, Matsumoto Y, Hosoya S (2004) Isolation of lignin
carbohydrate bonds in wood. Model experiments and preliminary application to pine wood.
J Wood Sci 50:141150
Karlsson O, Lundquist K, Meuller S, Westlid K (1988) On the acidolytic cleavage of arylglycerol
-aryl ethers. Acta Chem Scand B 42:4851
Keskar SS, Edye LA, Doherty WOS, Bartley JP (2011) The chemistry of acid catalyzed deligni-
fication of sugarcane bagasse in the ionic liquid trihexyl tetradecyl phosphonium chloride. J
Wood Chem Technol 32:7181
Kim H, Ralph J (2010) Solution-state 2D NMR of ball-milled plant cell wall gels in DMSO-d6/
pyridine-d5. Org Biomol Chem 8:576591
Kim Y, Mosier NS, Ladisch MR, Ramesh Pallapolu V, Lee YY, Garlock R et al (2011)
Comparative study on enzymatic digestibility of switchgrass varieties and harvests pro-
cessed by leading pretreatment technologies. Bioresour Technol 102:1108911096
Kimon KS, Leslie Alan E, William Orlando Sinclair D (2011) Enhanced saccharification kinetics
of sugarcane bagasse pretreated in 1-butyl-3-methylimidazolium chloride at high tempera-
ture and without complete dissolution. Bioresour Technol 102:93259329
King AWT, Parviainen A, Karhunen P, Matikainen J, Hauru LKJ, Sixta H et al (2012) Relative
and inherent reactivities of imidazolium-based ionic liquids: the implications for lignocel-
lulose processing applications. RSC Adv 2:80208026
Kishimoto T, Chiba W, Saito K, Fukushima K, Uraki Y, Ubukata M (2009) Influence of syrin-
gyl to guaiacyl ratio on the structure of natural and synthetic lignins. J Agric Food Chem
58:895901
Knill CJ, Kennedy JF (2003) Degradation of cellulose under alkaline conditions. Carbohydr
Polym 51:281300
Koo B-W, Min B-C, Gwak K-S, Lee S-M, Choi J-W, Yeo H et al (2012) Structural changes in
lignin during organosolv pretreatment of Liriodendron tulipifera and the effect on enzy-
matic hydrolysis. Biomass Bioenerg 42:2432
Kumar R, Mago G, Balan V, Wyman CE (2009) Physical and chemical characterizations of
corn stover and poplar solids resulting from leading pretreatment technologies. Bioresour
Technol 100:39483962
Larsen SU, Bruun S, Lindedam J (2012) Straw yield and saccharification potential for ethanol in
cereal species and wheat cultivars. Biomass Bioenerg 45:239250
Lawoko M, Henriksson G, Gellerstedt G (2006) Characterisation of lignin-carbohydrate com-
plexes (LCCs) of spruce wood (Picea abies L.) isolated with two methods. Holzforschung
60:156161
Le Ngoc Huyen T, Rmond C, Dheilly RM, Chabbert B (2010) Effect of harvesting date
on the composition and saccharification of Miscanthus x giganteus. Bioresour Technol
101:82248231
Lee C, Teng Q, Huang W, Zhong R, Ye Z-H (2009) Down-regulation of PoGT47C expression in
poplar results in a reduced glucuronoxylan content and an increased wood digestibility by
cellulase. Plant Cell Physiol 50:10751089
Li W, Sun N, Stoner B, Jiang X, Lu X, Rogers RD (2011) Rapid dissolution of lignocellulosic
biomass in ionic liquids using temperatures above the glass transition of lignin. Green Chem
13:20382047
Li X, Ximenes E, Kim Y, Slininger M, Meilan R, Ladisch M et al (2010) Lignin monomer com-
position affects Arabidopsis cell-wall degradability after liquid hot water pretreatment.
Biotechnol Biofuel 3(27):17
Liang H, Frost CJ, Wei X, Brown NR, Carlson JE, Tien M (2008) Improved sugar release from
lignocellulosic material by introducing a tyrosine-rich cell wall peptide gene in poplar.
CLEANSoil, Air, Water 36:662668
Lindedam J, Andersen SB, DeMartini J, Bruun S, Jrgensen H, Felby C et al (2012) Cultivar
variation and selection potential relevant to the production of cellulosic ethanol from wheat
straw. Biomass Bioenerg 37:221228
Lindedam J, Bruun S, Jorgensen H, Felby C, Magid J (2010) Cellulosic ethanol: Interactions
between cultivar and enzyme loading in wheat straw processing. Biotechnol Biofuel
3(25):110
Lionetti V, Francocci F, Ferrari S, Volpi C, Bellincampi D, Galletti R et al (2010) Engineering the
cell wall by reducing de-methyl-esterified homogalacturonan improves saccharification of
plant tissues for bioconversion. PNAS 107:616621
Lundquist K (1973) Acid degradation of lignin. Part VIII. Low molecular weight phenols from
acidolysis of birch lignin. Acta Chem Scand 27:25972606
Lundquist K, Lundgren R (1972) Acid degradation of lignin. Part VII. The cleavage of ether
bonds. Acta Chem Scand 26:20052023
MacFarlane DR, Pringle JM, Johansson KM, Forsyth SA, Forsyth M (2006) Lewis base ionic
liquids. Chem Commun, 19051917, doi: 10.1039/B516961P
Maloney MT, Chapman TW, Baker AJ (1985) Dilute acid hydrolysis of paper birch: kinetics
studies of xylan and acetyl-group hydrolysis. Biotechnol Bioeng 27:355361
Maloney VJ, Mansfield SD (2010) Characterization and varied expression of a membrane-bound
endo--1,4-glucanase in hybrid poplar. Plant Biotechnol J 8:294307
McGee JK, April GC (1982) Chemicals from renewable resources: hemicellulose behavior dur-
ing organosolv delignification of southern yellow pine. Chem Eng Commun 19:4956
Mechin V, Argillier O, Rocher F, Hebert Y, Mila I, Pollet B et al (2005) In search of a maize ideo-
type for cell wall enzymatic degradability using histological and biochemical lignin charac-
terization. J Agric Food Chem 53:58725881
Moller I, Srensen I, Bernal AJ, Blaukopf C, Lee K, bro J et al (2007) High-throughput mapping
of cell-wall polymers within and between plants using novel microarrays. Plant J 50:11181128
Morreel K, Dima O, Kim H, Lu F, Niculaes C, Vanholme R et al (2010) Mass spectrometry-
based sequencing of lignin oligomers. Plant Physiol 153:14641478
Muhammad N, Omar WN, Man Z, Bustam MA, Rafiq S, Uemura Y (2012) Effect of ionic liquid
treatment on pyrolysis products from bamboo. Ind Eng Chem Res 51:22802289
Ong RG (2011) Interactions between biomass feedstock characteristics and bioenergy produc-
tion: from the landscape to the molecular scale, PhD., Michigan State University, USA
Pattathil S, Avci U, Baldwin D, Swennes AG, McGill JA, Popper Z et al (2010) A comprehensive
toolkit of plant cell wall glycan-directed monoclonal antibodies. Plant Physiol 153:514525
Pauly M, Hake S, Kraemer FJ (2011) Maize variety and method of production. US Patent
13/152,219, Filed 2 Jun 2011
Pedersen M, Meyer AS (2010) Lignocellulose pretreatment severityrelating pH to biomatrix
opening. New Biotechnol 27:739750
Pordesimo LO, Hames BR, Sokhansanj S, Edens WC (2005) Variation in corn stover composi-
tion and energy content with crop maturity. Biomass Bioenerg 28:366374
Ralph J, Brunow G, Boerjan W (2007) Lignins. eLS. Wiley, pp 110, doi:
10.1002/9780470015902.a0020104
Rencoret J, Gutirrez A, Nieto L, Jimnez-Barbero J, Faulds CB, Kim H et al (2011) Lignin
composition and structure in young versus adult Eucalyptus globulus plants. Plant Physiol
155:667682
Riedlberger P, Weuster-Botz D (2012) New miniature stirred-tank bioreactors for parallel study
of enzymatic biomass hydrolysis. Bioresour Technol 106:138146
Rock K, Thelemann R, Jung H-J, Tschirner U, Sheaffer C, Johnson G (2009) Variation due to
growth environment in alfalfa yield, cellulosic ethanol traits, and paper pulp characteristics.
Bioenerg Res 2:7989
Rollin JA, Zhu Z, Sathitsuksanoh N, Zhang YHP (2011) Increasing cellulose accessibility is
more important than removing lignin: a comparison of cellulose solvent-based lignocellu-
lose fractionation and soaking in aqueous ammonia. Biotechnol Bioeng 108:2230
Saake B, Lehnen R (2007) Lignin. In: Ullmanns encyclopedia of industrial chemistry. Wiley-
VCH, doi: 10.1002/14356007.a15_305.pub3
Sanjaya Durrett TP, Weise SE, Benning C (2011) Increasing the energy density of vegetative tis-
sues by diverting carbon from starch to oil biosynthesis in transgenic Arabidopsis. Plant
Sannigrahi P, Ragauskas AJ, Miller SJ (2009) Lignin structural modifications resulting from eth-
anol organosolv treatment of loblolly pine. Energ Fuel 24:683689
Santoro N, Cantu S, Tornqvist C-E, Falbel T, Bolivar J, Patterson S et al (2010) A high-through-
put platform for screening milligram quantities of plant biomass for lignocellulose digest-
ibility. Bioenerg Res 3:93102
Scheller HV, Ulvskov P (2010) Hemicelluloses. Annu Rev Plant Biol 61:263289
Selig M, Tucker M, Law C, Doeppke C, Himmel M, Decker S (2011) High throughput determi-
nation of glucan and xylan fractions in lignocelluloses. Biotechnol Lett 33:961967
Selig MJ, Tucker MP, Sykes RW, Reichel KL, Brunecky R, Himmel ME et al (2010)
Lignocellulose recalcitrance screening by integrated high-throughput hydrothermal pretreat-
ment and enzymatic saccharification. Ind Biotechnol 6:104111
Sewalt VJH, Fontenot JP, Allen VG, Glasser WG (1996) Fiber composition and in vitro digest-
ibility of corn stover fractions in response to ammonia treatment. J Agric Food Chem
44:31363142
Shen H, He X, Poovaiah CR, Wuddineh WA, Ma J, Mann DGJ et al (2012) Functional charac-
terization of the switchgrass (Panicum virgatum) R2R3-MYB transcription factor PvMYB4
for improvement of lignocellulosic feedstocks. New Phytol 193:121136
Shi J, Ebrik MA, Yang B, Garlock RJ, Balan V, Dale BE et al (2011) Application of cellulase and
hemicellulase to pure xylan, pure cellulose, and switchgrass solids from leading pretreat-
ments. Bioresour Technol 102:1108011088
Shoseyov O, Shani Z, Levy I (2006) Carbohydrate binding modules: biochemical properties and
novel applications. Microbiol Mol Biol R 70:283295
Simmons BA, Loqu D, Ralph J (2010) Advances in modifying lignin for enhanced biofuel pro-
duction. Curr Opin Plant Biol 13:312319
Singh S, Simmons BA, Vogel KP (2009) Visualization of biomass solubilization and cellulose
regeneration during ionic liquid pretreatment of switchgrass. Biotechnol Bioeng 104:6875
Siqueira G, Milagres A, Carvalho W, Koch G, Ferraz A (2011) Topochemical distribution of
lignin and hydroxycinnamic acids in sugar-cane cell walls and its correlation with the enzy-
matic hydrolysis of polysaccharides. Biotechnol Biofuel 4:7
Slocombe SP, Cornah J, Pinfield-Wells H, Soady K, Zhang Q, Gilday A et al (2009) Oil accumu-
lation in leaves directed by modification of fatty acid breakdown and lipid synthesis path-
ways. Plant Biotechnol J 7:694703
Stewart JJ, Akiyama T, Chapple C, Ralph J, Mansfield SD (2009) The effects on lignin structure
of overexpression of ferulate 5-hydroxylase in hybrid poplar. Plant Physiol 150:621635
Sticklen M (2006) Plant genetic engineering to improve biomass characteristics for biofuels.
Curr Opin Biotechnol 17:315319
Stone B (2005) Cellulose: Structure and distribution. eLS. Wiley, pp 18, doi: 10.1038/npg.
els.0003892
Studer M, Brethauer S, DeMartini J, McKenzie H, Wyman C (2011a) Co-hydrolysis of hydro-
thermal and dilute acid pretreated populus slurries to support development of a high-
throughput pretreatment system. Biotechnol Biofuel 4(19):110
Studer MH, DeMartini JD, Brethauer S, McKenzie HL, Wyman CE (2010) Engineering of a
high-throughput screening system to identify cellulosic biomass, pretreatments, and enzyme
formulations that enhance sugar release. Biotechnol Bioeng 105:231238
Studer MH, DeMartini JD, Davis MF, Sykes RW, Davison B, Keller M et al (2011b) Lignin con-
tent in natural Populus variants affects sugar release. PNAS 108:63006305
Sun L, Simmons BA, Singh S (2011) Understanding tissue specific compositions of bioenergy
feedstocks through hyperspectral raman imaging. Biotechnol Bioeng 108:286295
Sun Y, Cheng JJ (2005) Dilute acid pretreatment of rye straw and bermudagrass for ethanol pro-
duction. Bioresour Technol 96:15991606
Tadesse H, Luque R (2011) Advances on biomass pretreatment using ionic liquids: an overview.
Energ Environ Sci 4:39133929
Tarkow H, Feist WC (1969) A mechanism for improving the digestibility of lignocellulosic mate-
rials with dilute alkali and liquid ammonia. In: Hajny GJ, Reese ET (eds) Cellulases and
their applications. American Chemical Society, Washington, D. C., pp 197218
Teymouri F, Alizadeh H, Laureano-Prez L, Dale B, Sticklen M (2004) Effects of ammonia fiber
explosion treatment on activity of endoglucanase from acidothermus cellulolyticus in trans-
genic plant. Appl Biochem Biotechnol 116:11831191
Tunc MS, van Heiningen ARP (2008) Hemicellulose extraction of mixed southern hardwood
with water at 150C: effect of time. Ind Eng Chem Res 47:70317037
Vrnai A, Siika-aho M, Viikari L (2010) Restriction of the enzymatic hydrolysis of steam-pre-
treated spruce by lignin and hemicellulose. Enzym Microb Technol 46:185193
Verma D, Kanagaraj A, Jin S, Singh ND, Kolattukudy PE, Daniell H (2010) Chloroplast-derived
enzyme cocktails hydrolyse lignocellulosic biomass and release fermentable sugars. Plant
Vitz J, Erdmenger T, Haensch C, Schubert US (2009) Extended dissolution studies of cellulose in
imidazolium based ionic liquids. Green Chem 11:417424
Voelker SL, Lachenbruch B, Meinzer FC, Kitin P, Strauss SH (2011) Transgenic poplars with
reduced lignin show impaired xylem conductivity, growth efficiency and survival. Plant,
Cell Environ 34:655668
Wallace G, Russell WR, Lomax JA, Jarvis MC, Lapierre C, Chesson A (1995) Extraction of phe-
nolic-carbohydrate complexes from graminaceous cell walls. Carbohydr Res 272:4153
Wang PY, Bolker HI, Purves CB (1964) Ammonolysis of uronic ester groups in birch xylan. Can
J Chem 42:24342439
Whitney SEC, Brigham JE, Darke AH, Reid JSG, Gidley MJ (1998) Structural aspects of the inter-
action of mannan-based polysaccharides with bacterial cellulose. Carbohydr Res 307:299309
Willfr S, Sundberg A, Hemming J, Holborn B (2005a) Polysaccharides in selected industrially
important softwood species. In: 59th appita annual conference and exhibition. Auckland,
New Zealand, pp 415422
Willfr S, Sundberg A, Pranovich A, Holborn B (2005b) Polysaccharides in some industrially
important hardwood species. Wood Sci Technol 39:601617
Wilson JR, Hatfield RD (1997) Structural and chemical changes of cell wall types during stem
development: consequences for fibre degradation by rumen microflora. Aust J Agric Res
48:165180
Wiman M, Dienes D, Hansen MAT, van der Meulen T, Zacchi G, Lidn G (2012) Cellulose
accessibility determines the rate of enzymatic hydrolysis of steam-pretreated spruce.
Wu H, Mora-Pale M, Miao J, Doherty TV, Linhardt RJ, Dordick JS (2011) Facile pretreatment
of lignocellulosic biomass at high loadings in room temperature ionic liquids. Biotechnol
Bioeng 108:28652875
Wyman CE, Dale BE, Balan V, Elander RT, Holtzapple MT, Ramirez RS, et al (2013)
Comparative performance of leading pretreatment technologies for biological conver-
sion of corn stover, poplar wood, and switchgrass to sugars, 1st edn. In: Wyman CE (ed)
Aqueous Pretreatment of Plant Biomass for Biological and Chemical Conversion to Fuels
and Chemicals. Wiley, London, pp 245265
Xiao L-P, Shi Z-J, Xu F, Sun R-C (2013) Characterization of lignins isolated with alkaline etha-
nol from the hydrothermal pretreated Tamarix ramosissima. Bioenerg Res 6:114
Xu H, Pan W, Wang R, Zhang D, Liu C (2012) Understanding the mechanism of cellulose disso-
lution in 1-butyl-3-methylimidazolium chloride ionic liquid via quantum chemistry calcula-
tions and molecular dynamics simulations. J Comput Aided Mol Des 26:329337
Zeng M, Ximenes E, Ladisch MR, Mosier NS, Vermerris W, Huang C-P et al (2012) Tissue-
specific biomass recalcitrance in corn stover pretreated with liquid hot-water: enzymatic
hydrolysis (Part 1). Biotechnol Bioeng 109:390397
Zhang H, Fangel JU, Willats WGT, Selig MJ, Lindedam J, Jrgensen H, et al (2013a)
Assessment of leaf/stem ratio in wheat straw feedstock and impact on enzymatic conver-
sion. GCB Bioenergy, doi: 10.1111/gcbb.12060, (Ahead of Print)
Zhang H, Wu J, Zhang J, He J (2005) 1-allyl-3-methylimidazolium chloride room temperature
ionic liquid: a new and powerful nonderivatizing solvent for cellulose. Macromolecules
38:82728277
Zhang X, Yang W, Blasiak W (2011a) Modeling study of woody biomass: interactions of cellu-
lose, hemicellulose, and lignin. Energ Fuel 25:47864795
Zhang Y, Culhaoglu T, Pollet B, Melin C, Denoue D, Barrire Y et al (2011b) Impact of lignin
structure and cell wall reticulation on maize cell wall degradability. J Agric Food Chem
59:1012910135
Zhang Z, OHara IM, Doherty WOS (2013b) Effects of pH on pretreatment of sugarcane bagasse
using aqueous imidazolium ionic liquids. Green Chem 15:431438
Zhao X, Zhang L, Liu D (2012) Biomass recalcitrance. Part II: Fundamentals of different pre-
treatments to increase the enzymatic digestibility of lignocellulose. Biofuel Bioprod
Biorefin 6:561579
Chapter 15
Lignocellulosic Biorefineries:
Concepts and Possibilities
Kenneth F. Reardon
Abstract To date, research, development, and commercialization within the bioen-

ergy industry has focused on the production of biofuels, with any unconverted bio-
mass used for production of electricity, biogas, animal feed, or fertilizer. However,
both the economics and the environmental impacts of biofuel production could be
improved by developing processes to obtain a wider range of chemicals (with higher
value) from biomass. Example products range from commodity chemicals such as
dicarboxylic acids to nutraceuticals. In this article, the concept of a biorefinery will
be explored, especially in comparison to a petroleum refinery. Various products and
options to produce non-fuel chemicals from plants biomass are outlined. Such pro-
cesses would lead to a more diverse and sustainable biorefinery.
Keywords Biorefinery Biorenewables Biomass Lignocellulosic Cellulosic

Biofuels
15.1Introduction
In the past decade, the term biorefinery has increased dramatically in usage
(Fig. 15.1), and a Google search on this term now returns more than 655,000
results. Biorefineries have been the subject of a report from the US National
Research Council (2000) and numerous conferences, books, and review articles.
The general concept of a biorefinerya facility that produces not only a bio-
fuel but also other chemical products and power from biomassis intended to
draw an analogy to a petroleum refinery (here, petrorefinery will be used to as a
matched shorthand term). Both bio- and petrorefineries have chemically complex
K. F. Reardon(*)
Department of Chemical and Biological Engineering,
Colorado State University, Fort Collins, CO 80523-1370, USA
e-mail: Kenneth.Reardon@colostate.edu
256 K. F. Reardon
Fig.15.1PubMed search results for the number of publications with terms related to biorefiner-
ies in the article title or abstract. Publications with biorefinery or biorefineries are the most
numerous (left axis) but the use of bio-based and biorenewable is also growing rapidly
feedstocks and the capability of producing numerous products, which provide

important benefits to the overall refinery economics.
But to what extent is this analogy valid? The goal of this article is to consider that
question and to focus in particular on the options available for co-products (non-fuel
products) that could be produced in a biorefinery. Two general types of biorefiner-
ies will be considered: those based on thermochemical conversion (gasification or
pyrolysis) and those based biochemical conversion (pretreatment-deconstruction-fer-
mentation). Other platforms, especially those that combine thermochemical and bio-
logical conversion steps, are under development and some are moving rapidly toward
commercialization (e.g., Zeachems biologicalchemical ethanol process). However,
many of the same concepts will apply to those processes. The feedstock to be consid-
ered for the biorefineries is lignocellulosic biomass (e.g., grass, wood, corn stover).
15.2Petroleum Refineries
Petrorefineries began operation in the mid-1800s, primarily to process crude petro-

leum into kerosene for use in lamps and heaters. With the expansion of the automobile
industry, production shifted to emphasize gasoline and diesel as products. Today, many
different products are obtained from crude petroleum (Fig.15.2). Modern refineries
may be extremely large (>90,000m3/day crude oil processed), are highly integrated,
and rely extensively on thermal and chemical processes (Fig.15.3, Table15.1). Major
unit operations in petrorefineries include separation processes (e.g., desalting, distil-
lation, evaporation) and reaction processes (e.g., thermal, catalytic, and steam crack-
ing; catalytic reforming; isomerization; alkylation). Petrorefineries shift their product
output over the course of a year, primarily to respond to increased gasoline demands in
the summer and increased heating oil needs in the winter (Suenaga and Smith 2011).
15 Lignocellulosic Biorefineries: Concepts and Possibilities 257
Fig.15.2Products (gallons) made from a 42-gal barrel of crude oil (http://www.eia.doe.gov/

kids/energy.cfm?page=oil_home-basics)
Fig.15.3Simplified schematic of the separation and conversion processes in a petrorefinery

that result in a wide range of hydrocarbon products (http://www.eia.gov/kids/energy.cfm?page=oil_
home-basics)
15.3Biorefineries: Past and Present
The term biorefinery appears to have first been proposed in a journal article by
Lynd et al. (1999) in an article discussing the concept of biocommodity engineer-
ing. These authors outlined a vision for the evolution of biorefineries from facili-
ties producing only a few products to those generating many outputs:
Table15.1Generalized comparison of petroleum refineries and biorefineries
258
Petroleum refinery Biorefinery Comments

Feedstock Crude oil Lignocellulosic biomass
(grasses, wood)
Sugar/starch fraction of
biomass (maize, sugarcane)
Vegetable oil or tallow
Algal biomass
Municipal solid waste
Feedstock variability Moderate High Biomass feedstocks may range from
wood to municipal solid waste to
algal biomass
Unit operations: Distillation, evaporation, and Separation of biomass components Petroleum refineries use continuous
separation other thermal methods based possible with solvents processes
on volatility differences at early but rarely done
and late stages of processing
Separation of products by distillation Biorefineries generally use batch
and other thermal methods processes
Unit operations: Continuous-flow heterogeneous Biochemical: soluble enzyme- Petroleum refineries use continuous
reaction catalytic or thermal reactions catalyzed reactions and suspended processes
cell fermentations
Thermochemical: thermal decomposition Biorefineries generally use batch
reactions and heterogeneous catalytic processes
reactions
Number of unit High Low
operations per facility
Degree of process High Low

K. F. Reardon
integration
(continued)
Table15.1continued
Petroleum refinery Biorefinery Comments
Current product spectrum Many (>100) Few Typical corn grain and sugarcane etha-
nol biorefinery products are ethanol,
electricity, and distillers grains
Flexibility of product High Very low
spectrum
Size 10,000 to >80,000m3/d crude oil feed- 4001,200m3/d product (Renewable Fuels Processing capabilities vary widely
stock (Wikipedia 2013) Association 2013) for both types of refinery.
Biorefinery values shown are for
ethanol
15 Lignocellulosic Biorefineries: Concepts and Possibilities
259
260 K. F. Reardon
Biocommodity processes and products are often treated as though one or at best a few
of these products would be manufactured in a single plant. Although this approach may
be necessary initially to keep the scope of marketing, financing, and technology devel-
opment manageable for first-of-a-kind plants, a multiproduct biorefinery configuration
is likely to be more cost effective in the long term. Such an evolution would be simi-
lar to that experienced in the petroleum refining industry in which the initial focus on
production of primarily kerosene with little revenue from the remaining fraction of oil
ultimately gave way to integrated refineries that convert virtually all feedstock fractions
into a wide range of valuable products (Lynd et al. 1999).
This implied definition of a biorefinery as producing multiple products from

biomass was subsequently adopted by the biomass community, and sometimes
extended to include food products (Ohara 2003). The term integrated biore-
finery was subsequently introduced, with essentially the same meaning (US
Department of Energy 2013). Despite this goal of multiple products, current
biorefineries produce relatively few products: corn grain ethanol biorefineries
often produce only ethanol, electricity, and distillers grains, while those pro-
ducing commodity chemicals focus on a single product (e.g., lactic acid or
3-hydroxypropionic acid).
While the concept of converting a complex feedstock into numerous prod-
ucts is a valuable and important part of the refinery analogy, it is interesting to
compare petro- and biorefineries on other grounds (Table15.1). In particular, the
variability of biomass feedstocks that can be considered for a biorefinery (from
high oil content algal biomass to high carbohydrate content wood) is much
larger than for a petrorefinery, meaning that biorefineries most likely will need
to be designed for particular feedstock types. The size of the two types of biore-
finery is also a significant difference; petrorefineries have become extremely
large, regional operations, whereas current biorefineries are much smaller. The
smaller size of biorefineries is partly a matter of their stage of development, but
also reflects issues associated with the supply and cost of transportation of their
relatively low energy density biomass feedstock to the refinery gate.
There are two primary biorefinery platforms for lignocellulosic biomass,
based on the type of conversion process used:
Biological: Biomass is pretreated using a variety of chemical and thermal
methods (Blanch 2012; Chundawat et al. 2011) to reduce the particle size and
make the carbohydrate polymers more accessible to the subsequent enzymatic
depolymerization step to fermentable sugars. Yeasts and other microorganisms
are then used to ferment the sugars to biofuels (e.g., ethanol, butanol) and
other chemical products.
Thermochemical: A process involving exposure to high temperatures, typi-
cally pyrolysis or gasification, is used to convert lignocellulosic biomass
primarily into a liquid (pyrolysis oil) or synthesis gas (a mixture of carbon
monoxide, hydrogen, and carbon dioxide), respectively (Digman et al. 2009).
These are then processed further using catalyzed chemical reactions to form
various products (Zhou et al. 2011).
15.4Opportunities to Further Realize the Biorefinery

Concept
15.4.1General Considerations
The primary aspect of the biorefinery concept is the output of multiple products from
a single facility. In 2004, the US Department of Energy produced reports identify-
ing the non-fuel products that were determined to be the best targets for biorefineries
on the basis of the pathways for production and the market sizes (US Department of
Energy 2004). An updated version of this Top 10 list was published by Bozell and
Petersen (2010). The chemicals identified in these reports (Table15.2) have been the
focus of many research and development efforts toward the goal of biorefining.
One factor that makes biorefineries different than their petroleum counterparts is
that oil is first fractionated and then each of those fractions is chemically converted
to one or more products. In contrast, both biological and thermochemical biorefinery
types convert biomass to an intermediate form (sugars, pyrolysis oil, or synthesis gas)
before those compounds are fractionated and/or formed into the end products. This
suggests that there are two main options for generating a suite of biorefinery products:
Enable simultaneous production of products by incorporating a fractionation
step prior to conversion
Design biorefineries to yield different products sequentially with a flexible plat-
form (e.g., change catalysts or microorganisms, operate at different conditions).
Table15.2Top chemicals for production from biomass

Conversion type Biological conversion Thermochemical: Biological or
synthesis gas thermochemical
Reference US Department of US Department of Bozell and Petersen
Energy (2004) Energy (2004) (2010)
Chemicals 1,4-diacids (succinic, Hydrogen Ethanol
fumaric and malic)
2,5-furan Methanol Furans
Dicarboxylic acid Glycerol and derivatives
3-hydroxy propionic Biohydrocarbons
acid (isoprene)
Aspartic acid Lactic acid
Glucaric acid Succinic acid
Glutamic acid Hydroxypropionic acid/
aldehyde
Itaconic acid Levulinic acid
Levulinic acid Sorbitol
3-hydroxybutyrolactone Xylitol
Glycerol
Sorbitol
Xylitol/Arabinitol
262 K. F. Reardon
The first opportunity for fractionation to obtain more products from biomass is
independent of the type of biorefinery. Here, chemicals could be extracted directly
from the lignocellulosic biomass. In one of the few examples of this concept, hex-
ane has been used to extract a mixture of long-chain alcohols from sugar cane and
switchgrass (Ravindranath et al. 2009). This mixture, called policosanol, has been
shown to have cholesterol-lowering potency (McCarty 2002).
15.4.2Thermochemical Biorefinery
In the case of pyrolysis oil, fractionation as well as chemical reactions can be used
to form products. For example, Naik et al. (2010) evaluated supercritical CO2 for
this purpose and determined that certain high value products (furanoids, pyra-
noids, and bezenoids) could be separated from the pyrolysis oil. Upgrading reac-
tions (with hydrogen) are used to lower the oxygen content of the molecules in the
oil mixture (Zhang et al. 2005). Ultimately, pyrolysis oil must be further fraction-
ated or blended into streams in a petrorefinery to obtain specific products.
Because synthesis gas has been produced for many years, catalysts and pro-
cesses have been developed to convert this gas in one or more steps to a range
of products from hydrocarbons to organic acids and aldehydes. However, a 2004
report from the US Department of Energy on suggested products to make from
synthesis gas recommended only hydrogen and methanol because catalyst costs
for the other products were too high (Table15.2) (US Department of Energy
2004). Catalyst development for FischerTropsch and other reactions is an ongo-
ing research topic (Abell and Montan 2011) and thus new options may emerge.
Depending upon the operating conditions of a thermochemical biorefinery, a
significant amount of biochar may be produced (Many 2012). This carbonaceous
material has been shown to have significant benefits for soil fertility (Spokas et
al. 2012) and has also received considerable attention as a means of sequestering
atmospheric carbon (Meyer et al. 2011).
15.4.3Biochemical Biorefinery
While the metabolic capabilities of naturally occurring microorganisms has pro-

vided the ability to produce a range of fermentation products, the advent of meta-
bolic engineering and synthetic biology technologies has dramatically increased
the number of products that can be produced by microorganisms growing on sug-
ars (Steen et al. 2010; Zhang et al. 2011). This includes not only previously pro-
duced fuel molecules such as alcohols and fatty acids, but also alkanes (Schirmer
et al. 2010) and other hydrocarbons (Dugar and Stephanopoulos 2011; Jang et al.
2012). The same approaches have been used to engineer microorganisms for the
production of non-fuel molecules from biomass (Curran and Alper 2012; Du et al.
2011; Jarboe et al. 2010). In most instances, metabolic engineering strategies have
been implemented in the easily modified industrial strains of Escherichia coli and
Saccharomyces cerevisiae, but research has also targeted known bacterial ferment-
ers such as Clostridium sp. (Tracy et al. 2012).
With this increasing capacity to modify the metabolism of microorganisms
and to form almost any metabolic product, biorefinery designers must decide
between an organism optimized to produce the highest yield of only one molecule
(and thus the sequential approach to achieving a multiple product biorefinery) or
an organism that produces a suite of useful molecules but with lower concentra-
tions and yields of any one of them. The type and cost of the separation processes
involved will be critical in making this choice.
15.5Future Perspectives
Within the next two years, several commercial-scale cellulosic biorefineries will
begin production. This is an important stage in the development of this industry.
Relative to current petrorefineries, the product diversity is much smaller. However,
this is an expected starting point, and petrorefineries also began with a small prod-
uct portfolio. It is interesting to note the rapid increase in refereed journal publica-
tions referencing not only biorefinery but also the terms bio-based products
and especially biorenewables (Fig.15.1).
In the future, several developments can be expected as biorefining matures:
The implementation of new types of chemical conversions, such as hydrogen-
olysis (Ruppert et al. 2012) and aqueous-phase reforming (Huber and Dumesic
2006; Vispute and Huber 2009)
Combinations of thermochemical and biological (chemical catalysts for sugars
and polyols (Ruppert et al. 2012; Zhou et al. 2011), microbial fermentations of
syngas (Henstra et al. 2007; Munasinghe and Khanal 2010)
The development of metabolically engineered microorganisms that are robust in
an industrial setting and have higher yields of the desired products
The development of cost-effective processes to extract and purify high-value
nutraceuticals from biorefinery feedstocks
The development of continuous biorefinery processing over the current batch
process-dominated approach
Increased acceptability of bio-based products in the chemical marketplace.
References
Abell S, Montan D (2011) Exploring iron-based multifunctional catalysts for FischerTropsch

synthesis: a review. ChemSusChem 4:15381556
Blanch HW (2012) Bioprocessing for biofuels. Curr Opin Biotechnol 23:390395
264 K. F. Reardon
Bozell JJ, Petersen GR (2010) Technology development for the production of biobased products
from biorefinery carbohydrates-the US Department of Energys top 10 revisited. Green
Chem 12:539554
Chundawat SPS, Beckham GT, Himmel ME, Dale BE (2011) Deconstruction of lignocellulosic
biomass to fuels and chemicals. Annu Rev Chem Biomol Eng 2:121145
Curran KA, Alper HS (2012) Expanding the chemical palate of cells by combining systems biol-
ogy and metabolic engineering. Metab Eng 14:289297
Digman B, Joo HS, Kim D-S (2009) Recent progress in gasification/pyrolysis technologies for
biomass conversion to energy. Environ Prog Sust Energy 28:4751
Du J, Shao Z, Zhao H (2011) Engineering microbial factories for synthesis of value-added prod-
ucts. J Ind Microbiol Biotechnol 38:873890
Dugar D, Stephanopoulos G (2011) Relative potential of biosynthetic pathways for biofuels and
bio-based products. Nat Biotechnol 29:10741078
Henstra AM, Sipma J, Rinzema A, Stams AJM (2007) Microbiology of synthesis gas fermenta-
tion for biofuel production. Curr Opin Biotechnol 18:200206
Huber G, Dumesic J (2006) An overview of aqueous-phase catalytic processes for production of
hydrogen and alkanes in a biorefinery. Catal Today 111:119132
Jang Y-S, Park JM, Choi S, Choi YJ, Seung DY, Cho JH, Lee SY (2012) Engineering of micro-
organisms for the production of biofuels and perspectives based on systems metabolic engi-
neering approaches. Biotechnol Adv 30:9891000
Jarboe LR, Zhang X, Wang X, Moore JC, Shanmugam KT, Ingram LO (2010) Metabolic engi-
neering for production of biorenewable fuels and chemicals: contributions of synthetic biol-
ogy. J Biomed Biotechnol 2010:118
Lynd L, Wyman C, Gerngross T (1999) Biocommodity engineering. Biotechnol Prog 15:777793
Many JJ (2012) Pyrolysis for biochar purposes: a review to establish current knowledge gaps
and research needs. Environ Sci Technol 46:79397954
McCarty MF (2002) Policosanol safely down-regulates HMG-CoA reductasepotential as a com-
ponent of the esselstyn regimen. Med Hypotheses 59:268279
Meyer S, Glaser B, Quicker P (2011) Technical, economical, and climate-related aspects of bio-
char production technologies: a literature review. Environ Sci Technol 45:94739483
Munasinghe PC, Khanal SK (2010) Biomass-derived syngas fermentation into biofuels: opportu-
nities and challenges. Bioresour Technol 101:50135022
Naik S, Goud VV, Rout PK, Dalai AK (2010) Supercritical CO2 fractionation of bio-oil produced
from wheat-hemlock biomass. Bioresour Technol 101:76057613
National Research Council (2000) Biobased industrial products: research and commerciali-
zation priorities. The National Academies Press, Washington DC. http://www.nap.edu/
catalog.php?record_id=5295
Ohara H (2003) Biorefinery. Appl Microbiol Biotechnol 62:474477
Ravindranath SV, Uppugundla N, Lay JO, Clausen EC, Wilkins M, Ingraham RG, West C, Wu Y,
Carrier DJ (2009) Policosanol, alpha-tocopherol, and moisture content as a function of tim-
ing of harvest of switchgrass (Panicum virgatum L.). J Agric Food Chem 57:35003505
Renewable Fuels Association (2013) Biorefinery locations. http://www.ethanolrfa.org/bio-refin-
ery-locations/. Accessed 5 April 2013
Ruppert AM, Weinberg K, Palkovits R (2012) Hydrogenolysis goes bio: from carbohydrates and
sugar alcohols to platform chemicals. Angew Chem Int Ed Engl 51:25642601
Schirmer A, Rude MA, Li X, Popova E, del Cardayre SB (2010) Microbial biosynthesis of
alkanes. Science 329:559562
Spokas KA, Cantrell KB, Novak JM, Archer DW, Ippolito JA, Collins HP, Boateng AA, Lima
IM, Lamb MC, McAloon AJ, Lentz RD, Nichols KA (2012) Biochar: a synthesis of its agro-
nomic impact beyond carbon sequestration. J Environ Qual 41:973989
Steen EJ, Kang Y, Bokinsky G, Hu Z, Schirmer A, Mcclure A, del Cardayre SB, Keasling JD
(2010) Microbial production of fatty-acid-derived fuels and chemicals from plant biomass.
Nature 463:559562
Suenaga H, Smith A (2011) Volatility dynamics and seasonality in energy prices: implications for
crack-spread price risk. Energy J 32:2758
Tracy BP, Jones SW, Fast AG, Indurthi DC, Papoutsakis ET (2012) Clostridia: the importance of
their exceptional substrate and metabolite diversity for biofuel and biorefinery applications.
Curr Op Biotechnol 23:364381
US Department of Energy (2004) Top value added chemicals from biomass. Volume 1: results of
screening for potential candidates from sugars and synthesis gas
US Department of Energy (2013) Integrated biorefineries. http://www1.eere.energy.gov/biomass/
integrated_biorefineries.html. Accessed 5 April 2013
Vispute TP, Huber GW (2009) Production of hydrogen, alkanes and polyols by aqueous phase
processing of wood-derived pyrolysis oils. Green Chem 11:14331445
Wikipedia (2013) List of oil refineries. http://en.wikipedia.org/wiki/List_of_oil_refineries.
Accessed 5 April 2013
Zhang F, Rodriguez S, Keasling JD (2011) Metabolic engineering of microbial pathways for
advanced biofuels production. Curr Opin Biotechnol 22:775783
Zhang S, Yan Y, Li T, Ren Z (2005) Upgrading of liquid fuel from the pyrolysis of biomass.
Biores Technol 96:545550
Zhou C-H, Xia X, Lin C-X, Tong D-S, Beltramini J (2011) Catalytic conversion of lignocellu-
losic biomass to fine chemicals and fuels. Chem Soc Rev 40:55885617
Chapter 16
Catalytic Dehydration of Lignocellulosic
Derived Xylose to Furfural
Basudeb Saha, Nathan S. Mosier and Mahdi M. Abu-Omar
Abstract In this chapter we present different biorefinary strategies for the produc-
tion of Furfural, a top ten platform chemical for making next generation fine chem-
icals and liquid fuels. Several research articles have been published demonstrating
the production of furfural using homogeneous and heterogeneous catalysts in single
and biphasic solvent systems. This article summarizes the finding of the most recent
research articles with critical discussion on the factors that control the yield and
selectivity of furfural. Among several factors, special emphasis has been given on
the improvement of partition coefficient of biphasic solvent systems and the effect
of pore size of the heterogeneous catalyst in enhancing furfural yield and selectivity.
Catalytic dehydration of xylose and its isomer form has been exemplified with Lewis
and Brnsted acidic catalysts in understanding the mechanistic role of the individual
acid sites in improving furfural yields and minimizing by-products formation.
Keywords Furfural Biphasic solvent Isomerization Sustainable process

Liquid fuels
B. Saha M. M. Abu-Omar(*)
Brown Laboratory, Department of Chemistry and School of Chemical Engineering,
The Center for Catalytic Conversion of Biomass to Biofuels (C3Bio),
Purdue University, 560 Oval Drive, West Lafayette, IN 47907, USA
e-mail: mabuomar@purdue.edu
N. S. Mosier
School of Agricultural and Biological Engineering, The Center for Catalytic Conversion
of Biomass to Biofuels (C3Bio), and Laboratory of Renewable Resources Engineering,
Purdue University, 500 Central Drive, West Lafayette, IN 47907, USA
268 B. Saha et al.
Fig.16.1Structure of xylan showing glycosidic bonds of xylose monomers
16.1Introduction
Furfural is an important biomass derived platform chemical (Werpy and

Peterson 2004), with an annual production volume of more than 200,000tons
(Kamm et al.2006). The Quaker Oats Company commercialized a method for
the production of furfural by treating hulls with dilute sulphuric acid as early as
1921 (Brownlee and Miner1948). The general methodology of furfural production
involves hydrolysis of xylan (Fig.16.1), a polymer of xylose, which is present in
lignocellulosic biomass, followed by catalytic dehydration of xylose with homo-
geneous or heterogeneous acidic materials. Xylose, a C5 sugar unit of hemicellu-
lose, is the second-most abundant component of biomass after cellulose (Dumitriu
and Dekker2005). Therefore, utilization of this abundant renewable C5 sugars for
the production of useful chemicals and fuels via environmentally and economi-
cally viable process is considered as a sustainable remediation to address the con-
cern of diminishing petroleum reservoir, variability in fossil energy price and high
dependence on petroleum feedstock (Dodds and Gross2007).
The research trend of furfural production and exploration of its potential appli-
cations have received significant attention in recent years after publication of Top
Value Added Chemicals from Biomass by Werpy and Peterson(2004). Furfural
can be used as a precursor for several high value chemicals and biofuel. Candidates
include 2-methylfuran (2-MF), 2-methyltetrahydrofuran (2-MeTHF), furfural alcohol
(FA), ethyl lelulinate (EL), -valerolactone (gVL) and long chain hydrocarbons of
diesel fraction (Corma et al.2007; Dutta et al.2012). Recently, furfural hydrogenation
product, 2-MF, derived hydrocarbon blended gasoline has been tested for 90,000km
road trial with promising outcomes, which prompted initiation of commercial scale
production of liquid hydrocarbon from 2-MF (Lange et al.2012). A scheme for
furfural production and its potential applications is shown in Fig.16.2.
16.2Furfural Production in Monophasic Solvent
The catalytic conversion of xylose, xylan and pre-treated biomass substrates to fur-
fural in monophasic and biphasic solvent systems has been investigated by using
homogeneous and heterogeneous catalytic materials containing Lewis and Brnsted
16 Catalytic Dehydration of Lignocellulosic Derived Xylose to Furfural 269
Fig.16.2Furfural production and its utilization routes for chemicals and liquid fuels
acidic sites. Like the Quaker Oats process, the conversion of xylose to furfural with
mineral acid catalysts (HCl, H2SO4) in aqueous medium achieved a maximum of 50%
furfural yields (Sievers et al.2009; Weingarten et al.2010; Yemis and Mazza2011).
However, mineral acid catalyzed processes suffer from corrosiveness and environ-
mental issues associated with the use of strong acids. In this context, Lewis acidic
metal chloride catalysts are advantageous for furfural production via isomerisation
of xylose to xylulose intermediate. Binder et al. have shown that CrCl3 and CrCl2
catalysts are effective for xylose conversion in N,N-dimethylacetamide (DMA)-LiCl
solvent, giving a maximum of 56% furfural yield in 4h at 100C (Binder et al.2010).
Depolymerisation of Birch xylan is, however, a major challenge using this catalytic
system, resulting in poor furfural yield (15%). Although pre-treatment of xylan with
HCl or 1-ethyl-3-methylimidazolium chloride ([EMIM]Cl) IL (ionic liquid) shows a
significant improvement in xylan hydrolysis (77%), a proportional increase in fur-
fural yield was not observed in the subsequent dehydration step in DMA-LiCl. The
same catalytic system in pure [EMIM]Cl solvent produced 63% furfural from xylan
in shorter time (3min) when the reaction was carried out under microwave assisted
hearing at 100C (Zhang and Zhao2010). This method is also effective for intact
biomass corn stalk, rice straw and pine wood due to high dissolution of biomass
in imidazolium-based ionic liquids. Furfural yields from these biomass species
are in the range of 2331% based on their pentose composition by weight (von
Sivers and Zacchi1995; Liu and Wyman2005; Jin and Chen2007). Potential
drawback of the IL solvent is that ionic liquids are expensive and the separation of
furfural from high boiling point ILs is energy intensive. Besides cost, ILs tend to
deactivate by the water formed during the dehydration reaction. Xylose conversion
270 B. Saha et al.
has also been investigated in high boiling point organic solvents such as dimethylsul-
foxide (DMSO) using heteropolyacids (Dias et al.2005) and Nafion catalysis (Lam
et al.2011). Although these reusable catalysts give modest furfural yields in the range
of 5867mol%, selectivity of the desired product is an issue due to the formation
of undesired humin by-products via oligomerization between xylose and furfural
(Dee and Bell2011). Such humin formation has also been a challenge for acid-
catalyzed formation of furfural from xylose in 1-butyl-3-methylimidazolium chlo-
ride ([BMIM]Cl) using Brnsted acids such as H2SO4 (Sievers et al.2009). Besides
humin formation in DMSO, the high boiling point of this solvent also possess a chal-
lenge for cost-effective separation of furfural with high purity, and therefore, these
processes are economically unfavorable for larger scale commercial applications.
16.3Furfural Production in Biphasic Media
Because of aforementioned disadvantages of monophasic solvent systems using

high boiling point organic solvents or poor yield in pure aqueous medium, current
research effort of furfural production is directed towards utilization of biphasic reac-
tion systems in batch or continuous reactor. In case of biphasic system, aqueous or
modified aqueous solution is used as the reactive phase. The organic layer of the
biphasic system acts as an extracting solvent for continuous separation of furfural
into the organic phase immediately as its being formed in the reactive phase. Thus,
lower concentration of furfural in the aqueous phase limits its rehydration with water
and thereby improves furfural yields (Qi et al.2009). This method allows easy sepa-
ration and reusability of the reactive aqueous phase containing both homogeneous
and heterogeneous catalysts. The partition coefficient (R), which is the ratio of fur-
fural in the organic phase to that in the aqueous phase, is an important parameter in
determining the overall performance of the biphasic system. Higher partitioning of
furfural into the organic layer improves effective extraction and hence increases prod-
uct selectivity as well as yield. Besides the nature of the organic solvents in determin-
ing the partition coefficient, the presence of inorganic salt, e.g. NaCl, in the aqueous
phase also increase the R values due to salting-out effect (Eisen and Joffe1966;
Tan and Aravinth1999). vom Stein et al.(2011) developed a method for furfural
production by using a biphasic solvent system comprising an aqueous solution of
FeCl36H2O, NaCl and xylose as a reactive phase and biomass derived 2-methyl-
tetrahydrofuran (2-MeTHF) as the extracting organic phase. This method exhibited a
maximum 71% furfural yield with 98% extraction capability into the organic phase.
Besides pure xylose conversion, this biphasic system containing FeCl3 catalyst has
been shown to convert Beechwood biomass extracted non-purified xylose solution
(30wt%) to furfural at a production of rate of 3.5g furfural/h (Fig.16.3).
Similar to the iron system, the water-NaCl-THF biphasic medium is effec-
tive for AlCl36H2O catalyzed conversion of xylose, giving 75% furfural yield
in 5min under microwave assisted heating (Yang et al.2012a, b). The potential
of the combined AlCl3/water-NaCl-THF system has further demonstrated for
Fig.16.3Conversion of Beechwood biomass derived xylose to furfural in aqueous-2-MeTHF

biphasic medium using iron catalyst
Table16.1Furfural yields from various sources of lignocellulosic biomass

Biomass Temp (C) Time (min) Furfural (%) Xylose (%)
Corn stover 160 60 55 <1
Pinewood 160 60 38 <1
Switchgrass 160 60 56 1
Poplar 160 60 64 <1
Cellulose/xylan 160 60 66 0
Pinewood 180 30 61 <1
sustainable furfural production by converting lignocellulosic biomass (corn stover,

pinewood, switchgrass, and poplar) (Yang et al. 2012a, b) to 5566mol% furfural
(Table16.1) based on literature values (Lu and Mosier2007, 2008) of their pen-
tose contents by weight.
Another strategy for converting the hemicellulose fraction of biomass to furfural
utilizes a biomass derived solvent, 2-sec-butylphenol (SBP), as an extracting phase
and an acidic aqueous layer as the reactive phase (Grbz et al.2012). This biphasic
system containing mineral acid catalyst produce high concentrations of furfural
with maximum 78% yield. The use of SBP solvent having high R values (90 and
50 with and without NaCl saturation in the aqueous phase) is advantageous because
of (1) fast furfural extraction, (2) low amounts of mineral acids dissolve in SBP
thereby eliminating energy intensive separation of mineral acid from the product
layer, and (3) SBP can be derived from lignin. While the authors list SBP higher
boiling point than the furfural product as an advantage, this feature can be inter-
preted as a drawback because it requires intensive energy to remove the product via
distillation (boiling point of furfural 162C). Nevertheless, similar lignin derived
alkylphenols, e.g., eugenol, can be envisaged as effective extracting solvents for
furfural production. A recent study shows an efficient xylose and xylan conversion
process with 72% furfural yield using maleic acid catalyst in an aqueous medium
272 B. Saha et al.
at 200C (Kim et al.2012). The synergistic effect of KCl and KI salts results in a
higher furfural yield (88%) and selectivity (95%).
Besides homogeneous catalysts, several heterogeneous catalysts, including
Zr-P, SiO2-Al2O3, WOx/ZrO2, -Al2O3 and HY zeolite (Weingarten et al.2011),
Amberlyst-15 (Takagaki et al.2010), hydrotalcite (Tuteja et al.2012), tin-tung-
sten mixed oxide (Yamaguchi et al.2011), Nafion 117 (Lam et al.2011), and Sn-
beta zeolite (Choudhary et al.2011) have been employed for furfural production to
take advantages of their easy separation and recyclability properties. These stud-
ies reveal that catalysts with higher Lewis acid sites are most active. However,
catalyst pore confinement is found to have an adverse effect on furfural selectiv-
ity. Adsorptiondesorption studies and decomposition experiments with furfural
in aqueous solution have confirmed that HY zeolite causes furfural to irreversi-
bly adsorb inside the zeolite pores and oligomarize to form humin side-products.
Therefore, micropores containing catalysts may not be suitable for xylose dehy-
dration due to strong adsorption of the product inside the catalyst pore.
16.4Mechanistic Studies of Furfural Formation
Experimental evidence suggests that glucose dehydration to 5-hydrxymethylfurfural

(HMF) occurs via the isomerization of glucose to fructose (Binder and Raines2009;
Roman-Leshkov et al.2010; De et al.2011; Pagn-Torres et al.2012). The occur-
rence of similar isomerization for xylose to xylulose intermediate followed by
dehydration to furfural was first established by comparing the xylose and xylulose
conversion rates with a series of Lewis and Brnsted acidic catalysts (Choudhary
et al.2011). The results show that xylose does not react with Brnsted acidic
catalysts such as Amberlyst-15 or HCl; however, when xylulose is the reactant,
conversion is ~66%, and furfural yield is 24%. On the other hand, a catalytic
system containing both Brnsted (HCl) and Lewis acidic sites (Sn-beta) can
directly convert xylose to furfural with evidence of formation of xylulose and
lyxose as intermediates. This result supports a reaction network in which xylulose
dehydrates rapidly to furfural via Brnsted acid catalysis and xylose isomerizes
to xylulose with a Lewis acid catalyst advocating for dual acidic sites of a catalyst.
Formation of xylulose is a key step to furfural and requires either functional group
rearrangement or a change in configuration on the C1 and C2 carbon atoms.
Structural studies using X-ray absorption fine structure (EXAFS) spectroscopy
reveals that Sn is substituted in pairs on opposite sides of six-member rings,
i.e. uniform crystallographic location of Sn in the crystal structure that leads
to sites with uniform catalytic activity and high chemical selectivity (Fig.16.4)
(Bare et al.2005). The results of Sn-beta zeolite catalyzed process indicate that
the active site of the catalyst interacts with the carbonyl group of C1 and the adja-
cent hydroxyl group on C2. Kinetic studies of isomerization reactions indicate
that certain acids and metals are able to transfer the hydrogen directly through
a hydride shift between C2 and C1 (Collyer and Blow1990). Involvement of
Fig.16.4Plausible reaction mechanism for xylose conversion to furfural via xylulose intermediate
Fig.16.5Kinetic profile for

xylose conversion with AlCl3
catalyst in water-NaCl/THF
solvent (black square xylose,
black circle xylulose, black
up-pointing triangle furfural)
1,2-hydride transfer step in the isomerisation of xylose to xylulose is also consistent

with results from deuterium labeling experiments using CrCl3 as catalyst (Binder
et al.2010). Lewis acidity in the catalyst is essential to polarize the carbonyl group
in the ketone while coordinating both the alcohol and the ketone to facilitate a
hydride shift between the two carbons (Corma et al.2001).
AlCl3 6H2O catalyzed conversion of xylose in water-NaCl/THF biphasic
solvent system shows direct evidence for xylulose formation as an intermediate
species (Yang et al.2012a, b). The kinetic profile (Fig.16.5) of xylose conver-
sion reveals xylulose formation in the early stages of the reaction, suggesting
that isomerization of xylose to xylulose takes place during the course of the
274 B. Saha et al.
dehydration reaction. Under comparable reaction conditions, xylose conversion

under conventional heating required longer reaction times (20min) and afforded
lower selectivity (64%) for furfural along with cross-polymer humin by-product.
Kinetic study of xylose conversion with maleic acid as catalyst (Kim et al.2012)
revealed that xylose conversion rates are lower in aqueous medium, which may be
due to furfural acting as a Brnsted base and reacting with H3O+. Thus, total acid
concentration of the aqueous medium decreases and consequently the conversion
rate of xylose is retarded (Antan et al.1991).
16.5Conclusions
Furfural is a promising biomass derived platform chemical for the production of

high value chemicals and liquid fuels that are currently obtained from petroleum
feedstock. Development of effective catalytic and reaction systems to achieve high
yield and selectivity of furfural from lignocellulosic biomass containing abundant
xylose constitute a viable strategy for the modern biorefinery. Although several
improvements have been made in laboratory scale processes to achieve high furfural
yield and selectivity by developing effective catalytic system containing balanced
Lewis and Brnsted acidic sites and biphasic reaction media, high production cost of
furfural is still a challenge to its sustainable utilization for making other high value
chemicals and fuels on commercial scale. A combination of both chemistry and
engineering efforts to design more effective catalysts through clear understanding of
the exact roles of Lewis and Brnsted acidic sites in minimization of by-products
formation, improvement in furfural yield and selectivity, and cost-effective purifica-
tion of the desired product(s) are necessary to lower furfural production cost.
Acknowledgments The authors acknowledge financial support from the Center for direct
Catalytic Conversion of Biomass to Biofuels (C3Bio), an Energy Frontier Research Center
funded by the U.S. Department of Energy, Office of Science, and Office of Basic Energy
Sciences under Award Number DE-SC0000997.
References
Antan MJ, Leesomboon T, Mok WS, Richards GN (1991) Mechanism of formation of 2-furalde-
hyde from d-xylose. Carbohydr Res 217:7185
Bare SR, Kelly SD, Sinkler W, Low JJ, Modica FS, Valencia S, Corma A, Nemeth LT (2005)
Uniform catalytic site in Sn--Zeolite determined using x-ray absorption fine structure. J Am
Chem Soc 127:1292412932
Choudhary V, Pinar AB, Sandler SI, Vlachos DG, Lobo RF (2011) Xylose isomerization to xylu-
lose and its dehydration to furfural in aqueous media. ACS Catal 1:17241728
Binder JB, Raines RT (2009) Simple chemical transformation of lignocellulosic biomass into
furans for fuels and chemicals. J Am Chem Soc 131:19791985
Binder JB, Blank JJ, Cefali AV, Raines RT (2010) Synthesis of furfural from xylose and xylan.
Chem Sus Chem 3:12681272
Brownlee HJ, Miner CS (1948) Industrial development of furfural. Ind Eng Chem 40:201204
Collyer CA, Blow DM (1990) Observations of reaction intermediates and the mechanism of aldose-
ketose interconversion by D-xylose isomerase. Proc Natl Acad Sci USA 87:13621366
Corma A, Nemeth LT, Renz M, Valencia S (2001) Sn-zeolite beta as a heterogeneous chemose-
lective catalyst for BaeyerVilliger oxidations. Nature 412:423425
Corma A, Ilorra S, Velty A (2007) Chemical routes for the transformation of biomass into chemi-
cals. Chem Rev 107:24112502
De S, Dutta S, Saha B (2011) Microwave assisted conversion of carbohydrates and biopoly-
mers to 5-hydroxymethylfurfural with aluminium chloride catalyst in water. Green Chem
13:28592868
Dias AS, Pillnger M, Valente AA (2005) Liquid phase dehydration of D-xylose in the presence of
Keggin-type heteropolyacids. Appl Catal A Gen 285:126131
Dee SJ, Bell AT (2011) A study of the acid-catalyzed hydrolysis of cellulose dissolved in ionic
liquids and the factors influencing dehydration of glucose and the formation of humins.
Chem Sus Chem 4:11661173
Dodds D, Gross R (2007) Chemicals from biomass. Science 318:12501251
Dumitriu S, Dekker M (2005) Polysaccharides: structural diversity and functional versatility, 2nd
edn. Marcel Dekker, New York
Dutta S, De S, Saha B, Alam I (2012) Advanced in conversion of hemicellulosic biomass to fur-
fural and upgrading to biofuels. Catal Sci Tech 2:20252036
Eisen EO, Joffe J (1966) Salt effects in liquidliquid equilibria. J Chem Eng Data 11:480484
Grbz EI, Wettstein SG, Dumesic JA (2012) Conversion of hemicellulose to furfural and lev-
ulinic acid using biphasic reactors with alkylphenol solvents. Chem Sus Chem 5:383387
Jin S, Chen H (2007) Fractionation of fibrous fraction from steam-exploded rice straw. Process
Biochem 42:188192
Kamm B, Gruber PR, Kamm M (2006) Biorefineriesindustrial processes and products. Wiley,
Weinheim
Kim ES, Liu S, Abu-Omar MM, Mosier NS (2012) Selective conversion of biomass hemicellu-
lose to furfural using maleic acid with microwave heating. Energy Fuels 26:12981304
Lam E, Majid E, Leung ACW, Cong JH, Mahmoud KA, Luong JHT (2011) Synthesis of furfural
from xylose by heterogeneous and reusable nafion catalysts. Chem Sus Chem 4:535541
Lange JP, Heide E, van der Buijtenen JV, Trice R (2012) Furfurala promising platform for lig-
nocellulosic biofuels. Chem Sus Chem 5:150166
Liu C, Wyman EE (2005) Partial flow of compressed hot water through corn stover to enhance hemi-
celluloses sugar recovery and enzymatic digestibility of cellulose. Biores Technol 96:19781986
Lu YL, Mosier NS (2007) Biomimetic catalysis for hemicellulose hydrolysis in corn stover.
Biotechnol Prog 23:116123
Lu YL, Mosier NS (2008) Kinetic modeling analysis of maleic acid-catalyzed hemicellulose
hydrolysis in corn stover. Biotechnol Bioeng 101:11701181
Pagn-Torres YJ, Wang T, Gallo JMR, Shanks BH, Dumesic JA (2012) Production of 5-hydroxy-
methylfurfural from glucose using a combination of Lewis and Brnsted acid catalysts in
water in a biphasic reactor with an alkylphenol solvent. ACS Catal 2:930934
Qi X, Watanabe M, Aida TM, Smith RL Jr (2009) Efficient process for conversion of fructose to
5-hydroxymethylfurfural with ionic liquids. Green Chem 11:13271331
Weingarten R, Tompsett GA, Conner WC Jr, Huber GW (2011) Design of solid acid catalysts
for aqueous phase dehydration of carbohydrates: the role of Lewis and Brnsted acid sites. J
Catal 279:174182
Roman-Leshkov Y, Moliner M, Labinger JA, Davis ME (2010) Mechanism of glucose isomeriza-
tion using a solid Lewis acid catalyst in water. Angew Chem Int Ed 49:89548957
Sievers C, Musin I, Marzialetti T, Valenzuela Olarte MB, Agrawal PK, Jones CW (2009) Acid-
catalyzed conversion of sugars and furfurals in an ionic-liquid phase. Chem Sus Chem 2:665671
Takagaki A, Ohara M, Nishimura S, Ebitani K (2010) One-pot formation of furfural from xylose
via isomerization and successive dehydration reactions over heterogeneous acid and base cat-
alysts. Chem Lett 39:838840
276 B. Saha et al.
Tan TC, Aravinth S (1999) Liquid-liquid equilibria of water/acetic acid/1-butanol systemeffects

of sodium (potassium) chloride and correlations. Fluid Phase Equilib 163:243257
Tuteja J, Nishimura S, Ebitani K (2012) One-pot synthesis of furans from various saccharides
using a combination of solid acid and base catalysts. Bull Chem Soc Jpn 85:275281
vom Stein T, Grande PM, Leitner W, de Maria PD (2011) Iron-catalyzed furfural production in
biobased biphasic systems: from pure sugars to direct use of crude xylose effluents as feed-
stock. Chem Sus Chem 4:15921594
von Sivers M, Zacchi G (1995) A techno-economical comparison of three processes for the pro-
duction of ethanol from pine. Biores Technol 51:4352
Weingarten R, Cho J, Conner WC Jr, Huber GW (2010) Kinetics of furfural production by dehy-
dration of xylose in a biphasic reactor with microwave heating. Green Chem 12:14231429
Werpy T, Peterson G (2004) Top value added chemicals from biomass, vol. 1. Pacific Northwestern
National Laboratory
Yamaguchi K, Sakurada T, Ogasawara Y, Mizuno N (2011) Tin-tungsten mixed oxide as effi-
cient heterogeneous catalyst for conversion of saccharides to furan derivatives. Chem Lett
40:542543
Yang Y, Hu CW, Abu-Omar MM (2012a) Conversion of carbohydrates and lignocellulosic bio-
mass into 5-hydroxymethylfurfural using AlCl3 6H2O catalyst in a biphasic solvent system.
Green Chem 14:509513
Yang Y, Hu C-W, Abu-Omar M-M (2012b) Synthesis of furfural from xylose, xylan, and biomass
using AlCl3 6H2O in biphasic media via xylose isomerization to xylulose. Chem Sus Chem
5:405410
Yemis O, Mazza G (2011) Acid-catalyzed conversion of xylose, xylan and straw into furfural by
microwave-assisted reaction. Biores Technol 102:73717378
Zhang Z, Zhao ZK (2010) Microwave-assisted conversion of lignocellulosic biomass into furans
in ionic liquid. Biores Technol 101:11111114
Chapter 17
Catalytic Oxidation of Lignin for the
Production of Low Molecular Weight
Aromatics
Joseph J. Bozell
AbstractLignin offers a number of attractive features as a starting material for

chemical production, and the only large scale source of aromatic moieties in nature.
It is highly abundant, comprising 1525wt% of lignocellulosic feedstocks, such as
agricultural materials or forest resources, making it the second most available source
of renewable carbon after cellulose.
17.1Introduction
Lignin offers a number of attractive features as a starting material for chemical produc-
tion, and the only large scale source of aromatic moieties in nature (Bozell et al. 2007).
It is highly abundant, comprising 1525wt% of lignocellulosic feedstocks, such as
agricultural materials or forest resources, making it the second most available source
of renewable carbon after cellulose. However, lignin faces significant disadvantages
as a chemical feedstock, with the primary drawback being a high level of structural
heterogeneity that arises from two primary sources. First, the biosynthesis of lignin
introduces heterogeneity as lignin is manufactured in the plant cell wall (Fig.17.1)
(Boerjan et al. 2003).
Lignin biosynthesis uses the three primary monolignols, p-hydroxycinnamyl
alcohol, coniferyl alcohol and sinapyl alcohol, leading to the well-recognized
p-hydroxyphenyl, guaiacyl and syringyl units, respectively, in the resulting lignin
biopolymer. During biosynthesis, these monolignols are converted into highly
delocalized phenoxy radicals that undergo radicalradical coupling and conver-
sion to the lignin polymer. Softwoods are primarily constructed of guaiacyl units,
while hardwoods contain significant amounts of both guaiacyl and syringyl units.
J. J. Bozell(*)
Center for Renewable Carbon, Center for the Catalytic Conversion of Biomass (C3Bio),
University of Tennessee, Knoxville, TN 37996, USA
e-mail: jbozell@utk.edu
278 J. J. Bozell
Fig.17.1Overview of lignin biosynthesis
Fig.17.2Suggested structure of native poplar lignin
Herbaceous feedstocks (grasses) offer additional complexity, as their structure

includes extensive crosslinking between lignin and hemicelluloses through ferulate
ester linkages. Moreover, the ferulates themselves undergo dimerization and trim-
erization, and can be introduced into the bulk lignin. The result of these coupling
reactions is the production of lignin as a complex aromatic biopolymer (Fig.17.2,
illustrating hardwood poplar) (Vanholme et al. 2010).
Figure17.2 also identifies many of the well-recognized substructural units that com-
prise the lignin polymer as it is found in the plant. However, using lignin as a chemical
17 Catalytic Oxidation of Lignin 279
feedstock requires its isolation from the lignocellulosic matrix, leading to the second
source of lignins structural heterogeneity. Any process used to isolate lignin inevita-
bly induces structural changes in the native material through loss of some substructural
units and introduction of new interunit linkages. As an example, switchgrass sam-
ples subjected to organosolv fractionation showed a marked change in the concentra-
tion of -O-4 units as the severity of the fractionation increased (Bozell et al. 2011b).
Quantitative 13C analysis of the NMR spectral region between 88 and 77ppm identifies
lignins -O-4 units, which can be 50% or greater of the interunit linkages in native
lignin. After organosolv fractionation, NMR spectra indicate that it is possible to nearly
eliminate all of these linkages at high severity (160C, 0.1M acid catalyst). Other
pretreatment processes result in similar changes. Dilute acid pretreatment can reduce
-O-4 units by 36% (Samuel et al. 2010), and steam explosion can nearly eliminate
them under proper conditions (Li et al. 2009). In parallel, lignins structure realizes an
increase in the number of free phenolic OH groups as the -O-4 units are cleaved.
Ongoing biorefinery development is providing access to high purity lignin as
new, inexpensive sources of renewable carbon through technology such as orga-
nosolv fractionation (Bozell et al. 2011a). Despite these structural challenges,
all lignin, regardless of source, comprises a network of electron rich aromatic
rings. Accordingly, such systems would be anticipated to undergo a wide range
of oxidation processes. Development of new oxidation catalysts would ideally
employ environmentally benign terminal oxidants such as O2 or HOOH, could
be adapted to operation in aqueous media, and would demonstrate reactivity
designed for the substructural units present in lignin as isolated by the biore-
finery. This chapter provides a brief overview of work in both nonselective and
selective lignin oxidation, and efforts within our group to develop selective cata-
lytic oxidation processes for lignin and lignin models.
17.2Nonselective Catalytic Lignin Oxidation
Lignin oxidation is a critical component of the kraft process used in the manufac-
ture of pulp and paper. The kraft industry is the largest current producer of lignin,
through the generation of about 120106 tons of pulp in 2005, leading to the paral-
lel production of about 72106t of lignin (Auhorn and Niemela 2007). Further,
kraft cellulose pulp retains a small amount of recalcitrant lignin that must be
removed chemically. Historically, however, the industry has not viewed this lignin
as a potential source of chemicals, and instead uses a wide range of nonselective
oxidation processes for its consumption. The simplest oxidation of lignin is its com-
bustion as fuel for the kraft operations chemical recovery boiler, a vital component
of a mills heat and energy balance (Fengel and Wegener 1984). In parallel, residual
lignin in kraft cellulose not extracted during pulping is subjected to nonselective
bleaching processes for its rapid oxidative deconstruction and removal. Industrially,
ClO2 is extensively used as a primary oxidant, but well recognized bleaching pro-
cesses employing O2, HOOH, O3 or sequential combinations of these oxidants are
also employed (Dence and Reeve 1996).
280 J. J. Bozell
Although ClO2 is an exceptionally efficient stoichiometric oxidant for lignin,

continuing industrial interest in eliminating all chlorine-containing reagents from
industrial bleaching has prompted evaluation of metal catalyzed approaches for
the activation of O2 or HOOH to minimize or replace ClO2. Extensive work has
been reported on nonselective catalytic oxidation of monomeric, dimeric and oli-
gomeric lignin models (Collinson and Thielemans 2010). Oxidation methodology
includes the use of O2 with catalytic Co(II) or Mn(II) salts as one electron biomi-
metic oxidants of a series of lignin models under harsh conditions (Dicosimo and
Szabo 1988), Cu(bis-ortho-phenanthroline) (Sippola and Krause 2005), or Co and
Fe metalloporphyrins (Crestini et al. 2004; Cui and Dolphin 1995; Zhu and Ford
1993), in organic or aqueous medium for the oxidation of several lignin models
using O2, aerobic oxidation catalyzed by polyoxometalates (Evtuguin et al. 2000;
Weinstock et al. 1997), or HOOH oxidation catalyzed by methyltrioxorhenium.
(Crestini et al. 2006; Crestini et al. 2005). Biocatalytic oxidation of lignin models
using O2 and a variety of laccases or laccases in the presence of a mediator has
also been widely examined as a means for removal of residual lignin from cel-
lulose (Barreca et al. 2003; Crestini et al. 2003; Elegir et al. 2005; Lahtinen etal.
2009; Li et al. 1999; Rochefort et al. 2004), and has been compared to metal cata-
lyzed systems (Bohlin et al. 2005). Electrocatalytic processes that mimic the action
of laccase in laccase mediator systems have been examined for the conversion of
lignin model monomers and dimers (Rochefort et al. 2002).
Expansion of these model studies to isolated lignin or cellulose pulp has also been
carried out. An extensive study on lignin oxidation to substituted aromatic carboxylic
acids used a mixed Co/Mn/Zr/Br catalyst in HOAc, modeling well-established indus-
trial processes for the conversion of alkyl aromatic compounds to the correspond-
ing carboxylic acid (e. g., the aerobic oxidation of para-xylene to terephthalic acid).
Several types of lignin were examined with this catalyst system, and gave a maxi-
mum yield of 10.9% organic products from hardwood organosolv lignin, as a mix-
ture of vanillin, vanillic acid, syringaldehyde and syringic acid (Partenheimer 2009).
Polyoxometalates and metalloporphyrins have also been examined with lignin in the
development of chlorine-free bleaching processes (Perng et al. 1994; Voitl and von
Rohr 2008; Weinstock et al. 1996). Electrochemical processes have been used for
lignin oxidation. Electrolysis of commercial alkali lignin was carried out at Ru/V/
Ti/O electrodes in ionic liquids to give 36wt% of low molecular weight aromatic
products as a complex mixture of more than 10compounds (Reichert et al. 2012).
Oxidation of kraft lignin using IrO2 electrodes afforded vanillin and vanillic acid as
the primary products, but no yields were reported (Tolba et al. 2010).
17.3Oxidation of Lignin to Vanillin
Historically, interest in selective oxidation of lignin to discrete low molecular

weight compounds began with efforts to produce vanillin from lignosulfonates iso-
lated from the sulfite pulping process. A much smaller amount of lignin (about
5.7 106t in 2005) is available from the sulfite process, but in contrast to the
kraft process, sulfite pulping generally does not use the lignin as fuel, affording
lignosulfonates as a separate process stream. The production of vanillin from
lignin in the pulp and paper industry has a long history, with an initial observation
of vanillin in lignin wastes appearing in 1875, and the first commercial scale pro-
duction in the US starting in 1936 (Hocking 1997). The process has typically been
carried out by alkaline oxidation of lignosulfonates isolated from sulfite pulping
of wood at high temperature using air as the oxidant. Yields of lignin are quite
low (510%), but the pulp industry has employed the process for the purpose of
generating additional revenue from pulping operations. Although 85% of vanillin
today is now produced from guaiacol, (da Silva et al. 2009) production from lignin
is still carried out on a limited scale by industry, in particular, Borregard (Voitl and
von Rohr 2010). The primary disadvantage to production of lignin from lignosul-
fonates is the cost associated with disposal of a large amount of residual lignin
waste after vanillin recovery. These costs led to elimination of nearly all commer-
cial lignin-based vanillin operations by the early 1990s (Hocking 1997).
Efforts have been made to improve this process through development of new oxi-
dation catalysts. It has long been recognized that the yields of vanillin from sulfite
liquors can be increased by the addition of metal catalysts such as Cu(II). Alkaline
oxidation of sulfite waste liquor gave 22% vanillin in the presence of CuSO4 5H2O,
and 13.5% yield from sulfite waste liquor solids (Pearl 1942). Alternatively, alka-
line nitrobenzene can be used as the oxidizing agent. Nitrobenzene oxidation of
spruce wood meal gave a 23% yield of vanillin (Creighton et al. 1941). However,
the requirement for stoichiometric amounts of nitrobenzene in the oxidation sig-
nificantly increases the cost of the process, making it unsuitable for industrial use.
Development of a process catalytic in nitrobenzene was attempted, by combining
nitrobenzene and Cu(II) catalyzed oxidations. However, the yield of vanillin using
this process was never greater than 6% (Bjorsvik 1999). As a result, the only remain-
ing commercial process for conversion of lignin to vanillin still uses aerobic alkaline
oxidation. Organosolv lignins have also been reported as starting materials for vanil-
lin synthesis. Organosolv lignin from eucalyptus, sugarcane bagasse and softwood via
the Acetosolv or Organocell process was converted to vanillin by oxidation in HOAc
with O2 catalyzed by Co(OAc)2. The optimum yield reported for any of these systems
was less than 6% (Goncalves and Schuchardt 1999).
The much more abundant kraft lignin can also be converted to vanillin, although
the yields remain uniformly low. Oxidation of black liquor from kraft pulping of
Pinus pilaster with oxygen in alkaline medium at~135C gave a maximum vanil-
lin yield of 1.2g/l, or about 0.9g of vanillin/100g of contained black liquor solids
after 50min. Comparative biocatalytic oxidation with Acinetobacter anitratus was
carried out, but gave a tenfold lower production of vanillin than chemical oxida-
tion (Mathias et al. 1995). Catalyzed and uncatalyzed oxidation of isolated kraft
lignin from eucalyptus with oxygen, or oxygen and added Cu(II) or Co(II) cata-
lysts gave<5% yield of low molecular weight products as a mixture of materials
(Villar et al. 2001). Engineering process analysis of this process to evaluate yield
and kinetics of the oxidation has been reported (Araujo et al. 2010). Oxidation
282 J. J. Bozell
of kraft lignin under acidic conditions was carried out using O2 as the oxidant in
the presence of the polyoxometalate H3PMo12O40 as a catalyst in MeOH at 170
to give a mixture of vanillin and methyl vanillate in combined yields of 78%.
However, the isolated material also contained significant amounts of oligomeric
material, indicating that further purification of the reaction products would be nec-
essary to obtain pure monomers (Voitl and von Rohr 2010, 2008). Again, nitroben-
zene oxidation of kraft lignin improves vanillin production, with a 13% vanillin
yield reported from pine lignin (Mathias and Rodrigues 1995). Engineering studies
to develop reactor systems for the continuous production and purification of vanil-
lin from kraft lignin have been reported (da Silva et al. 2009).
17.4Selective Catalytic Oxidation of Lignin
The current interest in developing biorefineries that integrate biobased chemical

and biofuel production has dramatically expanded the availability of lignin beyond
its traditional position within the pulp and paper industry. Accordingly, there has
been a clear transition from simply removing lignin as an unwanted impurity to
recognizing lignin as an important source of renewable carbon. The growing inter-
est in lignin as a chemical feedstock is evidenced by several recent reports on cat-
alytic processes for oxidative cleavage of lignin and lignin models from groups
not traditionally associated with biomass conversion (Hanson et al. 2012; Nichols
et al. 2010; Sergeev and Hartwig 2011; Son and Toste 2010; Zakzeski et al. 2010).
Our own research in this field dates from early efforts to selectively convert
lignin to substituted anthraquinones as catalysts for cellulose pulp manufacture
(Dimmel et al. 1999; Dimmel and Bozell 1991). More generally, our approaches
target the primary unifying structural feature of lignin, its network of oxygenated
aromatic rings, and in particular, the oxidation of para-substituted phenolics to ben-
zoquinones and other low molecular weight aromatics. This approach is designed
to take advantage of the structural units present in isolated biorefinery lignin, for
example, the recognized increase in the concentration of free phenolic OH groups
that occurs when -O-4 interunit linkages are cleaved. Many oxidative reactions
exist for preparing benzoquinones from phenolics. However, the great majority of
the reported oxidations are performed on phenolics that have no substituent para to
the hydroxyl group of the phenol. In contrast, every phenolic unit in lignin contains
a para substituent that must be selectively cleaved to realize a successful synthesis
of benzoquinones or related simple aromatic compounds. We found that Co-Schiff
base complexes catalyze this oxidation with O2 as the terminal oxidant (Fig.17.3,
illustrated using syringyl alcohol as the substrate) (Bozell et al. 1995).
In the presence of oxygen and an external ligand (L; typically an aromatic base
such as pyridine or imidazole) Co(salen) forms intermediate superoxo complex 1, in
which the normally high reactivity of oxygen is mediated. Subsequent reaction with
a lignin model, such as syringyl alcohol, abstracts the phenolic hydrogen to gener-
ate an intermediate phenoxy radical that is ultimately converted to a benzoquinone,
Fig.17.3Co-Schiff based catalyzed oxidation of para-substituted phenolics to quinones
Fig.17.4Improved oxidation of guaiacyl models in the presence of a hindered base
specifically dimethoxybenzoquinone (DMBQ) for the process shown in Fig.17.3.

The yields of the reaction were highly dependent on the structure of the substrate.
Syringyl models afforded quinone in 7190% yield, however, the correspond-
ing guaiacyl models were much less reactive, giving monomethoxybenzoquinone
(MMBQ) in 1227% yield. Depending on the substituent para to the phenolic
hydroxyl group and the catalyst employed, benzylic oxidation leading to the pro-
duction of substituted benzaldehydes could be observed in yields of 4550%.
Nonetheless, effective use of biorefinery lignin requires the ability to oxidize both
syringyl and guaiacyl units in high yield, and the failure of this catalytic system with
guaiacyl models was a disadvantage. These observations were attributed to an ina-
bility of the catalyst system to abstract a hydrogen atom from guaiacyl models to
form the necessary phenoxy radical intermediate as shown in Fig.17.3. Examination
of Hammett + constants for structurally similar phenols revealed that the presence
or absence of a single methoxy group on the aromatic ring could have a dramatic
effect on the rate of hydrogen removal from the substrate. Subsequent work revealed
that the yield of quinone from the guaiacyl model vanillyl alcohol is significantly
improved by adding a sterically hindered base to oxidations catalyzed by Co(salen)
or the alternative Co-Schiff base catalyst Co(N-Me salpr) (2, Fig.17.4) (Cedeno and
Bozell 2012). Oxidation of vanillyl alcohol to MMBQ proceeded in over 50% yield
upon addition of diisopropylethylamine, diisopropylamine or triethylamine. In con-
trast, oxidations in the absence of these bases displayed a maximum 21% yield of
quinone. Importantly, addition of this hindered base did not reduce the high yields of
product observed for syringyl model oxidation.
Both NMR and UVVIS measurements indicate that the hindered base does
not coordinate to the Co catalysts and thus its role must involve other mechanistic
pathways (Fig.17.5).
284 J. J. Bozell
Fig.17.5Proposed mechanism for the oxidation of guaiacyl models in the presence of a hin-
dered base
Our proposed mechanism postulates an initial deprotonation of the sub-

strate by the sterically hindered base to form a more readily oxidized phenolate
anion (Rappoport 2003). Transformation of the phenolate to the phenoxy radical
occurs either with oxygen or the intermediate Co-superoxo complex, ultimately
leading to quinone formation. This effect is currently limited, as addition of hin-
dered bases to oxidations of other guaiacyl models such as isoeugenol, eugenol,
or vanillin and vanillin acetal gave only low yields (010%) of MMBQ. Further
investigation into this alternative mechanism and computational evaluation of the
intermediate Co complexes is currently underway.
Application of Co-Schiff base catalyzed oxidation to biorefinery lignin samples
also induces conversion to quinones and structurally related aromatics. Treating
mixtures of poplar and switchgrass lignin with Co(salen) and oxygen affords
approximately 10wt% yield of low molecular weight products, most of which
is DMBQ. 2D-HMQC NMR analysis resolves these materials from the residual
lignin present after the oxidation (Fig.17.6).
Interestingly, oxidations of lignin are not improved by the addition of either an
external basic ligand or a non-ligating hindered base, suggesting that the lignin itself
may be acting in those capacities. Although the yields are still low, they are equiva-
lent to similar lignin oxidation processes reported in the literature. Work is currently
underway to optimize our catalyst system for the production of these materials.
Because of the tendency of Co(salen)/O2 complexes to undergo deactivation dur-
ing oxidation (Busch 1988), we also investigated other species that contain an oxygen
centered free radical structurally analogous to Co-superoxo complexes and reported
that stoichiometric NO2 could be used for this conversion (Dimmel et al. 1996).
Subsequent work revealed that catalytic NO2 in the presence of O2 also converted
para-substituted phenols to benzoquinones (Bozell et al. 1998). In initial experi-
ments, syringyl alcohol was treated with a stoichiometric amount of NaNO2 and a
small amount of concentrated HNO3 or HCl (a convenient source of NO2) in MeOH
under argon at 20C to afford DMBQ in low yield. However, introduction of 1
atmosphere of O2 to this reaction has a dramatic effect, allowing isolation of DMBQ
in much higher yields (8090%) using only catalytic amounts of NaNO2 to produce
8090% yields of DMBQ from syringyl alcohol. With NaNO2 levels as low as 5%,
DMBQ was still formed in yields of 7075% (Fig.17.7).
The mechanism of NO2 oxidation displays similarities to that of Co-Schiff base/
O2 oxidations, but is complicated by the presence of additional oxides of nitrogen
Fig.17.6Typical 2D HMQC spectrum of products from Co-Schiff base/O2 oxidation
Fig.17.7Oxidation of lignin models using catalytic NO2 in O2
that can be formed under these conditions (Bosch et al. 1994). Kochi has shown that
NO2 oxidation of hydroquinone dialkyl ethers to quinones occurs via a radical cation
that results from the reaction of the substrate with the NO2 disproportionation prod-
uct NO+NO3 (Rathore et al. 1994). Thus, substrates used in our study are assumed
to undergo conversion to a radical cation upon reaction with NO+. Formation of the
quinone occurs via reaction of this cation with the nitrate counterion (Fig.17.8).
This sequence forms HONO that is converted back to NO2 via N2O3 formation
and subsequent O2 oxidation of N2O3 to N2O4, continuing the catalytic cycle. In par-
allel, we observe that direct reaction with NO2 affords ring nitration rather than oxida-
tion through formation of a phenoxy radical and subsequent trapping of that radical by
a second molecule of NO2. This mechanistic path is similar to that described for the
reaction of phenols with Co-Schiff base complexes and O2 (Nishinaga et al. 1981).
We have also examined this process for the oxidation of organosolv lignin samples,
and found that the yield of quinone is again fairly low, around 68%.
286 J. J. Bozell
Fig.17.8Proposed mechanistic steps in catalytic NO2 oxidation of para-substituted phenolics
17.5Conclusions
The ability to convert lignins heterogeneous structure to a single compound in

high yield remains a grand challenge to the effective use of lignocellulosic feed-
stocks within the biorefinery. However, the transition of lignin from a byproduct
largely sequestered within the pulp and paper industry to a material of general
availability for biorefining has driven development of new catalytic oxidations
designed to take advantage of the structural features resulting from its isolation.
With the growing interest in lignin as a chemical feedstock, new catalytic pro-
cesses offer the potential of improved conversion of this highly abundant material.
Acknowledgments This work was supported as part of the Center for Direct Catalytic
Conversion of Biomass to Biofuels (C3Bio), an Energy Frontier Research Center funded by the US
Department of Energy, Office of Science, Office of Basic Energy Sciences under Award Number
DE-SC0000997, and US Department of Energy Office of Industrial Technologies.
References
Araujo JDP, Grande CA, Rodrigues AE (2010) Vanillin production from lignin oxidation in a
batch reactor. Chem Eng Res Des 88:10241032
Auhorn WJ, Niemela K (2007) Process chemicals for the production of chemical pulp. Prof
Papermaking 2:1020
Barreca AM, Fabbrini M, Galli C, Gentili P, Ljunggren S (2003) Laccase/mediated oxidation of a
lignin model for improved delignification procedures. J Mol Catal B 26:105110
Bjorsvik HR (1999) Fine chemicals from lignosulfonates. 1. Synthesis of vanillin by oxidation of
lignosulfonates. Org Proc Res Devel 3:330340
Boerjan W, Ralph J, Baucher M (2003) Lignin biosynthesis. Ann Rev Plant Biol 54:519546
Bohlin C, Persson P, Gorton L, Lundquist K, Jonsson LF (2005) Product profiles in enzymic and
non-enzymic oxidations of the lignin model compound erythro-1-(3,4-dimethoxyphenyl)2-
(2-methoxyphenoxy)-1,3-propanediol. J Mol Catal B 35:100107
Bosch E, Rathore R, Kochi JK (1994) Novel catalysis of hydroquinone autoxidation with nitro-
gen-oxides. J Org Chem 59:25292536
Bozell JJ, Hames BR, Dimmel DR (1995) Cobalt-Schiff base complex-catalyzed oxidation of
para-substituted phenolicspreparation of benzoquinones. J Org Chem 60:23982404
Bozell JJ, Hoberg JO, Dimmel DR (1998) Catalytic oxidation of para-substituted phenols with
nitrogen dioxide and oxygen. Tetrahedron Lett 39:22612264
Bozell JJ, Holladay JE, Johnson D, White JF (2007) Top value added chemicals from biomass.
Volume IIresults of screening for potential candidates from Biorefinery Lignin. Pacific
Northwest National Laboratory, Richland, WA
Bozell JJ, Black SK, Myers M, Cahill D, Miller WP, Park S (2011a) Solvent fractionation of
renewable woody feedstocks: Organosolv generation of biorefinery process streams for the
production of biobased chemicals. Biomass Bioenergy 35:41974208
Bozell JJ, OLenick CJ, Warwick S (2011b) Biomass fractionation for the biorefinery: heteronu-
clear multiple quantum coherence-nuclear magnetic resonance investigation of lignin isolated
from solvent fractionation of switchgrass. J Agric Food Chem 59:92329242
Busch DH (1988) Synthetic dioxygen carriers for dioxygen transport. In: Martell AE, Sawyer DT
(eds) Oxygen complexes and oxygen activation by transition metals. Plenum, New York
Cedeno D, Bozell JJ (2012) Catalytic oxidation of para-substituted phenols with cobalt-Schiff
base complexes/O2-selective conversion of syringyl and guaiacyl lignin models to benzoqui-
nones. Tetrahedron Lett 53:23802383
Collinson SR, Thielemans W (2010) The catalytic oxidation of biomass to new materials focus-
ing on starch, cellulose and lignin. Coord Chem Rev 254:18541870
Creighton RHJ, McCarthy JL, Hibbert H (1941) Aromatic aldehydes from spruce and maple
woods. J Am Chem Soc 63:312
Crestini C, Jurasek L, Argyropoulos DS (2003) On the mechanism of the laccase-mediator sys-
tem in the oxidation of lignin. Chem Eur J 9:53715378
Crestini C, Pastorini A, Tagliatesta P (2004) Metalloporphyrins immobilized on motmoril-
lonite as biomimetic catalysts in the oxidation of lignin model compounds. J Mol Catal A
208:195202
Crestini C, Pro P, Neri V, Saladino R (2005) Methyltrioxorhenium: a new catalyst for the activa-
tion of hydrogen peroxide to the oxidation of lignin and lignin model compounds. Biorg Med
Chem 13:25692578
Crestini C, Caponi MC, Argyropoulos DS, Saladino R (2006) Immobilized methyltrioxo rhenium
(MTO)/H2O2 systems for the oxidation of lignin and lignin model compounds. Biorg Med
Chem 14:52925302
Cui F, Dolphin D (1995) Iron porphyrin catalyzed oxidation of lignin model compounds: Oxidation
of phenylpropane and phenylpropene model compounds. Can J Chem 73:21532157
da Silva EAB, Zabkova M, Araujo JD, Cateto CA, Barreiro MF, Belgacem MN, Rodriques AE
(2009) An integrated process to produce vanillin and lignin-based polyurethanes from kraft
lignin. Chem Eng Res Des 87:12761292
Dence CW, Reeve DW (1996) Pulp bleachingprinciples and practice. TAPPI, Atlanta
Dicosimo R, Szabo HC (1988) Oxidation of lignin model compounds using single-electron-
transfer catalysts. J Org Chem 53:16731679
Dimmel DR, Bozell JJ (1991) Pulping catalysts from lignin. Tappi J 74:239241
Dimmel DR, Karim MR, Savidakis MC, Bozell JJ (1996) Pulping catalysts from lignin. 5.
Nitrogen dioxide oxidation of lignin models to benzoquinones. J Wood Chem Technol
16:169189
Dimmel DR, Althen E, Savidakis M, Courchene C, Bozell JJ (1999) New quinone-based pulping
catalysts. Tappi J 82:8389
Elegir G, Daina S, Zoia L, Bestetti G, Orlandi M (2005) Laccase mediator system: oxidation of
recalcitrant lignin model structures present in residual kraft lignin. Enzyme Microb Technol
37:340346
Evtuguin DV, Daniel AID, Silvestre AJD, Amado FML, Neto CP (2000) Lignin aerobic oxida-
tion promoted by molybdovanadophosphate polyanion PMo7V5O40 (8). Study on the oxi-
dative cleavage of beta-O-4 aryl ether structures using model compounds. J Mol Catal A
154:217224
288 J. J. Bozell
Fengel D, Wegener G (1984) Wood: Chemistry, ultrastructure, reactions. Walter de Gruyter,

Berlin
Goncalves AR, Schuchardt U (1999) Oxidation of organosolv lignins in acetic acidInfluence of
oxygen pressure. Appl Biochem Biotechnol 779:127132
Hanson SK, Wu R, Silks LAP (2012) CC or CO bond cleavage in a phenolic lignin model
compound: selectivity depends on vanadium catalyst. Angew Chem Int Ed 51:34103413
Hocking MB (1997) Vanillin: synthetic flavoring from spent sulfite liquor. J Chem Educ
74:10551059
Lahtinen M, Kruus K, Boer H, Kemell M, Andberg M, Viikari L, Sipila J (2009) The effect of
lignin model compound structure on the rate of oxidation catalyzed by two different fungal
laccases. J Mol Catal B 57:204210
Li KC, Xu F, Eriksson KEL (1999) Comparison of fungal laccases and redox mediators in oxida-
tion of a nonphenolic lignin model compound. Appl Environ Microbiol 65:26542660
Li JB, Gellerstedt G, Toven K (2009) Steam explosion lignins; their extraction, structure and
potential as feedstock for biodiesel and chemicals. Bioresour Technol 100:25562561
Mathias AL, Rodrigues AE (1995) Production of vanillin by oxidation of pine kraft lignins with
oxygen. Holzforschung 49:273278
Mathias AL, Lopretti MI, Rodrigues AE (1995) Chemical and biological oxidation of pinus-pin-
aster lignin for the production of vanillin. J Chem Technol Biotechnol 64:225234
Nichols JM, Bishop LM, Bergman RG, Ellman JA (2010) Catalytic CO bond cleavage of 2-ary-
loxy-1-arylethanols and its application to the depolymerization of lignin-related polymers. J
Am Chem Soc 132:1255412555
Nishinaga A, Tomita H, Nishizawa K, Matsuura T, Ooi S, Hirotsu K (1981) Regioselective for-
mation of peroxyquinolatocobalt(III) complexes in the oxygenation of 2,6-di-tert-butylphe-
nols with cobalt(II) Schiff-base complexes. J Chem Soc Dalton, pp 15041514
Partenheimer W (2009) The aerobic oxidative cleavage of lignin to produce hydroxyaromatic
benzaldehydes and carboxylic acids via metal/bromide catalysts in acetic acid/water mix-
tures. Adv Synth Catal 351:456466
Pearl IA (1942) Vanillin from lignin materials. J Am Chem Soc 64:14291431
Perng YS, Oloman CW, Watson PA, James BR (1994) Catalytic oxygen bleaching of wood pulp
with metal porphyrin and phthalocyanine complexes. Tappi J 77:119125
Rappoport Z (2003) The chemistry of phenols. Wiley, Hoboken
Rathore R, Bosch E, Kochi JK (1994) Selective nitration versus oxidative dealkylation of hydro-
quinone ethers with nitrogen-dioxide. Tetrahedron 50:67276758
Reichert E, Wintringer R, Volmer DA, Hempelmann R (2012) Electro-catalytic oxidative cleav-
age of lignin in a protic ionic liquid. Phys Chem Chem Phys 14:52145221
Rochefort D, Bourbonnais R, Leech D, Paice MG (2002) Oxidation of lignin model compounds by
organic and transition metal-based electron transfer mediators. Chem Commun pp 11821183
Rochefort D, Leech D, Bourbonnais R (2004) Electron transfer mediator systems for bleaching
of paper pulp. Green Chem 6:1424
Samuel R, Pu YQ, Raman B, Ragauskas AJ (2010) Structural characterization and comparison
of switchgrass ball-milled lignin before and after dilute acid pretreatment. Appl Biochem
Biotechnol 162:6274
Sergeev AG, Hartwig JF (2011) Selective, nickel-catalyzed hydrogenolysis of aryl ethers.
Science 332:439443
Sippola VO, Krause AOI (2005) Bis(o-phenanthroline)copper-catalysed oxidation of lignin
model compounds for oxygen bleaching of pulp. Catal Today 100:237242
Son S, Toste FD (2010) Non-oxidative vanadium-catalyzed CO bond cleavage: application to
degradation of lignin model compounds. Angew Chem Int Ed 49:37913794
Tolba R, Tian M, Wen JL, Jiang ZH, Chen AC (2010) Electrochemical oxidation of lignin at
IrO2-based oxide electrodes. J Electroanal Chem 649:915
Vanholme R, Demedts B, Morreel K, Ralph J, Boerjan W (2010) Lignin biosynthesis and struc-
ture. Plant Physiol 153:895905
Villar JC, Caperos A, Garcia-Ochoa F (2001) Oxidation of hardwood kraft-lignin to phenolic

derivatives with oxygen as oxidant. Wood Sci Technol 35:245255
Voitl T, von Rohr PR (2008) Oxidation of lignin using aqueous polyoxometalates in the presence
of alcohols. ChemSusChem 1:763769
Voitl T, von Rohr PR (2010) Demonstration of a process for the conversion of kraft lignin into
vanillin and methyl vanillate by acidic oxidation in aqueous methanol. Ind Eng Chem Res
49:520525
Weinstock IA, Atalla RH, Reiner RS, Moen MA, Hammel KE, Houtman CJ, Hill CL (1996) A new
environmentally benign technology and approach to bleaching kraft pulp. Polyoxometalates for
selective delignification and waste mineralization. New J Chem 20:269275
Weinstock IA, Atalla RH, Reiner RS, Moen MA, Hammel KE, Houtman CJ, Hill CL, Harrup MK
(1997) A new environmentally benign technology for transforming wood pulp into paper
Engineering polyoxometalates as catalysts for multiple processes. J Mol Catal A 116:5984
Zakzeski J, Jongerius AL, Weckhuysen BM (2010) Transition metal catalyzed oxidation of Alcell
lignin, soda lignin, and lignin model compounds in ionic liquids. Green Chem 12:12251236
Zhu WM, Ford WT (1993) Oxidation of lignin model compounds in water with dioxygen and
hydrogen-peroxide catalyzed by metallophthalocyanines. J Mol Catal 78:367378

Plants and Bioenergy

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Plants and Bioenergy

Uploaded by

Copyright:

Available Formats

Advances in Plant Biology 4

For further volumes:

Plants and BioEnergy

ISBN 978-1-4614-9328-0 ISBN 978-1-4614-9329-7 (eBook)

Library of Congress Control Number: 2013953273

Springer Science+Business Media New York 2014

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

A nations vision for developing renewable and sustainable energy resources is

In Part I, Social and Economic Impacts of a Bioenergy Agriculture, we

Part I Economics of Bioenergy

1 The Prospects of First Generation Ethanol

2 Can Energy Policy Drive Sustainable Land Use? Lessons

3 Advanced Biofuels: Economic Uncertainties, Policy Options,

4 Algae Farming and Its Bio-Products. . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

Part II Biomass Biology

5 Regional Gene Pools for Restoration, Conservation,

6 Mining Genetic Diversity of Sorghum as a Bioenergy Feedstock. . . . . 81

7 Genetics, Genomics and Crop Modelling: Integrative Approaches

8 Camelina: An Emerging Oilseed Platform for Advanced Biofuels

9 Perspectives in Brazil of the Contribution of Palm Trees

10 Xylan Biosynthesis in Plants, Simply Complex . . . . . . . . . . . . . . . . . . . 153

11 Towards Redesigning Cellulose Biosynthesis

Part III Biomass Processing

12 Developing Novel Enzyme Repertoires for the Efficient

13 Using Natural Plant Cell Wall Degradation Mechanisms

14 Linking Plant Biology and Pretreatment: Understanding

15 Lignocellulosic Biorefineries: Concepts and Possibilities . . . . . . . . . . . 255

16 Catalytic Dehydration of Lignocellulosic Derived Xylose

17 Catalytic Oxidation of Lignin for the Production of Low Molecular

Patricia Guardabassi and Jos Goldemberg

AbstractThere are great perspectives to the development of second-generation

Keywords Biofuels First generation Developing countries Barriers

M. C. McCann et al. (eds.), Plants and BioEnergy, Advances in Plant Biology 4, 3

The economic growth of developing countries and the maintenance of consump-

1.2The State of the Art of Ethanol Production

or second generation technologies, to produce biofuels include the enzymatic con-

1.3Existing Mandates to Ethanol Production and Use

Ethanol is produced in Africa to replace gasoline since de 1970s, in countries

In 2009, Mozambique Council of Ministers approved the National Biofuels

1.3.2Latin America and Caribbean

1.4Main Obstacles to the Development of Ethanol Industry

The development of a stable institutional and political environmental is required

Jeremy Woods and Nicole Kalas

The mandated increase in bioenergy as a means to decarbonise our energy supply,

M. C. McCann et al. (eds.), Plants and BioEnergy, Advances in Plant Biology 4, 13

2.1.1Changing Patterns in Global Energy Supplies

Fig.2.1Net imports of oil (20002035). Source IEA (2011)

2.1.2Future World Energy Production and Price Trends:

2.1.2.1World Energy Production and Price Trends

2.1.2.2Global Bioenergy Policy and Consumption

Bioenergy policies are motivated by climate change mitigation targets, energy

diesel=1.32 million litres).

Fig.2.4Global trends in fossil fuel prices (19702011). Source BP (2012)

5Mtoe=million tonnes of oil equivalent (1Mtoe=41.9PJ).

2.2What is Sustainable Bioenergy and What to Measure

Fig.2.6Measures of sustainable bioenergy

emissions (e.g., Searchinger 2010). Figure2.7 further illustrates the divergence

Fig.2.8Main drivers of large-scale land acquisitions. Source Land Matrix (2013)

2.2.1Sustainable Agricultural Intensification: The Future

The production of bioenergy sits within a larger system of agricultural produc-

B. Ensuring that there is adequate stability in food suppliesand protecting the

2.3Bioenergy and Land Use

2.3.1Technical Bioenergy Potentials and Global Land Availability

1.2The State of the Art of Ethanol Production

1.3Existing Mandates to Ethanol Production and Use

1.3.2Latin America and Caribbean

1.4Main Obstacles to the Development of Ethanol Industry

2.1.1Changing Patterns in Global Energy Supplies

2.1.2Future World Energy Production and Price Trends:

2.1.2.1World Energy Production and Price Trends

2.1.2.2Global Bioenergy Policy and Consumption

2.2What is Sustainable Bioenergy and What to Measure

2.2.1Sustainable Agricultural Intensification: The Future

2.3Bioenergy and Land Use

2.3.1Technical Bioenergy Potentials and Global Land Availability

2.3.2Energy Crop Yield Estimates

2.3.4Integrating Biomass Supply Chains and Sustainable

2.4How Can Bioenergy Policy Drive Sustainable

3.3Future Oil Price

3.4Feedstock Supply and Cost

4.3Commercial Uses of Algae

4.3.1Waste Water Treatment

4.3.3Food and Feed Products

4.5Summary and Conclusions

5.4Gene Flow Concerns and the Role of Unimproved

5.5Coexistence of Improved Varieties and Conservation

6.2Sorghum: Excellent Genetic and Genomic Resources

6.3Sorghum Has Been Identified as a Preferred Biomass

6.4Sweet Sorghum is Ideal for Both Sugar- and Cellulosic-

6.6Sweet Sorghum Does Not Compromise Food Security

6.7Sweet Sorghum is Ideal for Double Cropping

6.8Development of Perennial Sorghum

6.9Breeding for Cold Tolerance as a Mean to Extend

6.10Development of Herbicides Tolerance

6.11Breeding Strategies for Sweet Sorghum